diff --git a/.opencode/tool/github-triage.ts b/.opencode/tool/github-triage.ts index f25ca48b384b..e861e1e467b2 100644 --- a/.opencode/tool/github-triage.ts +++ b/.opencode/tool/github-triage.ts @@ -4,7 +4,7 @@ import { tool } from "@opencode-ai/plugin" const TEAM = { tui: ["kommander", "simonklee"], desktop_web: ["Hona", "Brendonovich"], - core: ["jlongster", "rekram1-node", "nexxeln", "kitlangton", "starptech"], + core: ["jlongster", "rekram1-node", "nexxeln", "kitlangton"], inference: ["fwang", "MrMushrooooom", "starptech"], windows: ["Hona"], } as const diff --git a/.termux-release-automation b/.termux-release-automation index 7960e0917e05..a9915909beb1 160000 --- a/.termux-release-automation +++ b/.termux-release-automation @@ -1 +1 @@ -Subproject commit 7960e0917e05f1de03e90a16785222c7156056e5 +Subproject commit a9915909beb119fefaab180912bafe04493e6998 diff --git a/artifacts/glm52-rise-video/sst-env.d.ts b/artifacts/glm52-rise-video/sst-env.d.ts new file mode 100644 index 000000000000..64441936d7a0 --- /dev/null +++ b/artifacts/glm52-rise-video/sst-env.d.ts @@ -0,0 +1,10 @@ +/* This file is auto-generated by SST. Do not edit. */ +/* tslint:disable */ +/* eslint-disable */ +/* deno-fmt-ignore-file */ +/* biome-ignore-all lint: auto-generated */ + +/// + +import "sst" +export {} \ No newline at end of file diff --git a/bun.lock b/bun.lock index 320f8544ef50..085654e942cd 100644 --- a/bun.lock +++ b/bun.lock @@ -29,7 +29,7 @@ }, "packages/app": { "name": "@opencode-ai/app", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@corvu/drawer": "catalog:", "@dnd-kit/abstract": "0.5.0", @@ -95,7 +95,7 @@ }, "packages/cli": { "name": "@opencode-ai/cli", - "version": "1.17.18", + "version": "1.17.19", "bin": { "lildax": "./bin/lildax.cjs", }, @@ -143,7 +143,7 @@ }, "packages/codemode": { "name": "@opencode-ai/codemode", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "acorn": "8.15.0", "effect": "catalog:", @@ -157,7 +157,7 @@ }, "packages/console/app": { "name": "@opencode-ai/console-app", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@cloudflare/vite-plugin": "1.15.2", "@ibm/plex": "6.4.1", @@ -193,7 +193,7 @@ }, "packages/console/core": { "name": "@opencode-ai/console-core", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@aws-sdk/client-sts": "3.782.0", "@jsx-email/render": "1.1.1", @@ -220,7 +220,7 @@ }, "packages/console/function": { "name": "@opencode-ai/console-function", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@ai-sdk/anthropic": "3.0.82", "@ai-sdk/openai": "3.0.48", @@ -242,7 +242,7 @@ }, "packages/console/mail": { "name": "@opencode-ai/console-mail", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@jsx-email/all": "2.2.3", "@jsx-email/cli": "1.4.3", @@ -266,7 +266,7 @@ }, "packages/console/support": { "name": "@opencode-ai/console-support", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@cloudflare/vite-plugin": "1.15.2", "@opencode-ai/console-core": "workspace:*", @@ -286,7 +286,7 @@ }, "packages/core": { "name": "@opencode-ai/core", - "version": "1.17.18", + "version": "1.17.19", "bin": { "opencode": "./bin/opencode", }, @@ -303,7 +303,7 @@ "@ai-sdk/google-vertex": "4.0.128", "@ai-sdk/groq": "3.0.31", "@ai-sdk/mistral": "3.0.27", - "@ai-sdk/openai": "3.0.53", + "@ai-sdk/openai": "3.0.84", "@ai-sdk/openai-compatible": "2.0.41", "@ai-sdk/perplexity": "3.0.26", "@ai-sdk/provider": "3.0.8", @@ -338,7 +338,7 @@ "drizzle-orm": "catalog:", "effect": "catalog:", "fuzzysort": "3.1.0", - "gitlab-ai-provider": "6.10.0", + "gitlab-ai-provider": "6.11.0", "glob": "13.0.5", "google-auth-library": "10.5.0", "gray-matter": "4.0.3", @@ -381,7 +381,7 @@ }, "packages/desktop": { "name": "@opencode-ai/desktop", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@zip.js/zip.js": "2.7.62", "effect": "catalog:", @@ -435,7 +435,7 @@ }, "packages/effect-drizzle-sqlite": { "name": "@opencode-ai/effect-drizzle-sqlite", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "drizzle-orm": "catalog:", "effect": "catalog:", @@ -449,7 +449,7 @@ }, "packages/effect-sqlite-node": { "name": "@opencode-ai/effect-sqlite-node", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "effect": "catalog:", }, @@ -461,7 +461,7 @@ }, "packages/enterprise": { "name": "@opencode-ai/enterprise", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@hono/standard-validator": "catalog:", "@opencode-ai/core": "workspace:*", @@ -493,7 +493,7 @@ }, "packages/function": { "name": "@opencode-ai/function", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@octokit/auth-app": "8.0.1", "@octokit/rest": "catalog:", @@ -509,7 +509,7 @@ }, "packages/http-recorder": { "name": "@opencode-ai/http-recorder", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@effect/platform-node": "4.0.0-beta.83", "@effect/platform-node-shared": "4.0.0-beta.83", @@ -540,7 +540,7 @@ }, "packages/llm": { "name": "@opencode-ai/llm", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@opencode-ai/schema": "workspace:*", "@smithy/eventstream-codec": "4.2.14", @@ -559,7 +559,7 @@ }, "packages/opencode": { "name": "opencode", - "version": "1.17.18", + "version": "1.17.19", "bin": { "opencode": "./bin/opencode", }, @@ -579,7 +579,7 @@ "@ai-sdk/google-vertex": "4.0.128", "@ai-sdk/groq": "3.0.31", "@ai-sdk/mistral": "3.0.27", - "@ai-sdk/openai": "3.0.53", + "@ai-sdk/openai": "3.0.84", "@ai-sdk/openai-compatible": "2.0.41", "@ai-sdk/perplexity": "3.0.26", "@ai-sdk/provider": "3.0.8", @@ -632,7 +632,7 @@ "drizzle-orm": "catalog:", "effect": "catalog:", "fuzzysort": "3.1.0", - "gitlab-ai-provider": "6.10.0", + "gitlab-ai-provider": "6.11.0", "glob": "13.0.5", "google-auth-library": "10.5.0", "gray-matter": "4.0.3", @@ -690,7 +690,7 @@ }, "packages/plugin": { "name": "@opencode-ai/plugin", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@ai-sdk/provider": "3.0.8", "@opencode-ai/sdk": "workspace:*", @@ -766,7 +766,7 @@ }, "packages/sdk/js": { "name": "@opencode-ai/sdk", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "cross-spawn": "catalog:", }, @@ -781,7 +781,7 @@ }, "packages/server": { "name": "@opencode-ai/server", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@opencode-ai/core": "workspace:*", "@opencode-ai/protocol": "workspace:*", @@ -796,7 +796,7 @@ }, "packages/session-ui": { "name": "@opencode-ai/session-ui", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@kobalte/core": "catalog:", "@opencode-ai/core": "workspace:*", @@ -840,7 +840,7 @@ }, "packages/slack": { "name": "@opencode-ai/slack", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@opencode-ai/sdk": "workspace:*", "@slack/bolt": "^3.17.1", @@ -853,7 +853,7 @@ }, "packages/stats/app": { "name": "@opencode-ai/stats-app", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@ibm/plex": "6.4.1", "@opencode-ai/stats-core": "workspace:*", @@ -886,7 +886,7 @@ }, "packages/stats/core": { "name": "@opencode-ai/stats-core", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@aws-sdk/client-athena": "3.933.0", "@planetscale/database": "1.19.0", @@ -905,7 +905,7 @@ }, "packages/stats/server": { "name": "@opencode-ai/stats-server", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@aws-sdk/client-firehose": "3.933.0", "@effect/platform-node": "catalog:", @@ -946,7 +946,7 @@ }, "packages/tui": { "name": "@opencode-ai/tui", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@opencode-ai/core": "workspace:*", "@opencode-ai/plugin": "workspace:*", @@ -973,7 +973,7 @@ }, "packages/ui": { "name": "@opencode-ai/ui", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@kobalte/core": "catalog:", "@pierre/diffs": "catalog:", @@ -1024,7 +1024,7 @@ }, "packages/web": { "name": "@opencode-ai/web", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@astrojs/cloudflare": "12.6.3", "@astrojs/markdown-remark": "6.3.1", @@ -3851,7 +3851,7 @@ "github-slugger": ["github-slugger@2.0.0", "", {}, "sha512-IaOQ9puYtjrkq7Y0Ygl9KDZnrf/aiUJYUpVf89y8kyaxbRG7Y1SrX/jaumrv81vc61+kiMempujsM3Yw7w5qcw=="], - "gitlab-ai-provider": ["gitlab-ai-provider@6.10.0", "", { "dependencies": { "@anthropic-ai/sdk": "^0.71.0", "@anycable/core": "^0.9.2", "graphql-request": "^6.1.0", "isomorphic-ws": "^5.0.0", "openai": "^6.16.0", "socket.io-client": "^4.8.1", "vscode-jsonrpc": "^8.2.1", "zod": "^3.25.76" }, "peerDependencies": { "@ai-sdk/provider": ">=3.0.0", "@ai-sdk/provider-utils": ">=4.0.0" } }, "sha512-oWEZ06rDO6JjB7INHO882wyBAQqCZVHiDHwCs5M+VPmdDj8TzhGXcYesA2CcV5RoI5lfHLKwGp5uKFB62VWpqw=="], + "gitlab-ai-provider": ["gitlab-ai-provider@6.11.0", "", { "dependencies": { "@anthropic-ai/sdk": "^0.71.0", "@anycable/core": "^0.9.2", "graphql-request": "^6.1.0", "isomorphic-ws": "^5.0.0", "openai": "^6.16.0", "socket.io-client": "^4.8.1", "vscode-jsonrpc": "^8.2.1", "zod": "^3.25.76" }, "peerDependencies": { "@ai-sdk/provider": ">=3.0.0", "@ai-sdk/provider-utils": ">=4.0.0" } }, "sha512-/LTD5qtSbGGnWd7/HB72YioSMNYqF7RlABojBqRqZ8UK5DpxoTp/dnT7K/tIScDWMYPGpJqX/B7JIOh0xeXFLQ=="], "glob": ["glob@13.0.5", "", { "dependencies": { "minimatch": "^10.2.1", "minipass": "^7.1.2", "path-scurry": "^2.0.0" } }, "sha512-BzXxZg24Ibra1pbQ/zE7Kys4Ua1ks7Bn6pKLkVPZ9FZe4JQS6/Q7ef3LG1H+k7lUf5l4T3PLSyYyYJVYUvfgTw=="], @@ -6023,7 +6023,7 @@ "@openauthjs/openauth/jose": ["jose@5.9.6", "", {}, "sha512-AMlnetc9+CV9asI19zHmrgS/WYsWUwCn2R7RzlbJWD7F9eWYUTGyBmU9o6PxngtLGOiDGPRu+Uc4fhKzbpteZQ=="], - "@opencode-ai/core/@ai-sdk/openai": ["@ai-sdk/openai@3.0.53", "", { "dependencies": { "@ai-sdk/provider": "3.0.8", "@ai-sdk/provider-utils": "4.0.23" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-Wld+Rbc05KaUn08uBt06eEuwcgalcIFtIl32Yp+GxuZXUQwOb6YeAuq+C6da4ch6BurFoqEaLemJVwjBb7x+PQ=="], + "@opencode-ai/core/@ai-sdk/openai": ["@ai-sdk/openai@3.0.84", "", { "dependencies": { "@ai-sdk/provider": "3.0.14", "@ai-sdk/provider-utils": "4.0.38" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-cmgbeJL0bbY0yTJH4/AdmP5E7MjWRL9G8UdhIi0JlV/So03o82ORJofW8OzwCZPTORVQblFbpZXYGDcUd9NdUQ=="], "@opencode-ai/core/@ai-sdk/openai-compatible": ["@ai-sdk/openai-compatible@2.0.41", "", { "dependencies": { "@ai-sdk/provider": "3.0.8", "@ai-sdk/provider-utils": "4.0.23" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-kNAGINk71AlOXx10Dq/PXw4t/9XjdK8uxfpVElRwtSFMdeSiLVt58p9TPx4/FJD+hxZuVhvxYj9r42osxWq79g=="], @@ -6391,7 +6391,7 @@ "opencode/@ai-sdk/cerebras": ["@ai-sdk/cerebras@2.0.60", "", { "dependencies": { "@ai-sdk/openai-compatible": "2.0.54", "@ai-sdk/provider": "3.0.12", "@ai-sdk/provider-utils": "4.0.33" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-Rnok3cThg6awBwaDSyiZpgRpbV7pqxGYrA89LODCo5cuEHeP2h0AM0lLHP7zIkclAdXfOm4wldKi/S2T/DGCOw=="], - "opencode/@ai-sdk/openai": ["@ai-sdk/openai@3.0.53", "", { "dependencies": { "@ai-sdk/provider": "3.0.8", "@ai-sdk/provider-utils": "4.0.23" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-Wld+Rbc05KaUn08uBt06eEuwcgalcIFtIl32Yp+GxuZXUQwOb6YeAuq+C6da4ch6BurFoqEaLemJVwjBb7x+PQ=="], + "opencode/@ai-sdk/openai": ["@ai-sdk/openai@3.0.84", "", { "dependencies": { "@ai-sdk/provider": "3.0.14", "@ai-sdk/provider-utils": "4.0.38" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-cmgbeJL0bbY0yTJH4/AdmP5E7MjWRL9G8UdhIi0JlV/So03o82ORJofW8OzwCZPTORVQblFbpZXYGDcUd9NdUQ=="], "opencode/@ai-sdk/openai-compatible": ["@ai-sdk/openai-compatible@2.0.41", "", { "dependencies": { "@ai-sdk/provider": "3.0.8", "@ai-sdk/provider-utils": "4.0.23" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-kNAGINk71AlOXx10Dq/PXw4t/9XjdK8uxfpVElRwtSFMdeSiLVt58p9TPx4/FJD+hxZuVhvxYj9r42osxWq79g=="], @@ -6899,6 +6899,10 @@ "@octokit/rest/@octokit/core/before-after-hook": ["before-after-hook@4.0.0", "", {}, "sha512-q6tR3RPqIB1pMiTRMFcZwuG5T8vwp+vUvEG0vuI6B+Rikh5BfPp2fQ82c925FOs+b0lcFQ8CFrL+KbilfZFhOQ=="], + "@opencode-ai/core/@ai-sdk/openai/@ai-sdk/provider": ["@ai-sdk/provider@3.0.14", "", { "dependencies": { "json-schema": "^0.4.0" } }, "sha512-5X1k57JBJ4H7H1QjX7CnJYAB1I19r/trVZTMcSms7/kLNZ8RaU4Nt2agcwZzv82Hfx6Q7/TOLU7agAKeFfc8cA=="], + + "@opencode-ai/core/@ai-sdk/openai/@ai-sdk/provider-utils": ["@ai-sdk/provider-utils@4.0.38", "", { "dependencies": { "@ai-sdk/provider": "3.0.14", "@standard-schema/spec": "^1.1.0", "eventsource-parser": "^3.0.8" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-/HHGmtKllqjg1OLc023v9w9kK3laW7Z6TzfZukYQWCsGBbzB9p60zTvvpXFVcs44NZBVXL3viOa1HRKUbeee8g=="], + "@opencode-ai/desktop/@actions/artifact/@actions/http-client": ["@actions/http-client@2.2.3", "", { "dependencies": { "tunnel": "^0.0.6", "undici": "^5.25.4" } }, "sha512-mx8hyJi/hjFvbPokCg4uRd4ZX78t+YyRPtnKWwIl+RzNaVuFpQHfmlGVfsKEJN8LwTCvL+DfVgAM04XaHkm6bA=="], "@opencode-ai/web/@shikijs/transformers/@shikijs/core": ["@shikijs/core@3.20.0", "", { "dependencies": { "@shikijs/types": "3.20.0", "@shikijs/vscode-textmate": "^10.0.2", "@types/hast": "^3.0.4", "hast-util-to-html": "^9.0.5" } }, "sha512-f2ED7HYV4JEk827mtMDwe/yQ25pRiXZmtHjWF8uzZKuKiEsJR7Ce1nuQ+HhV9FzDcbIo4ObBCD9GPTzNuy9S1g=="], @@ -7121,6 +7125,10 @@ "opencode/@ai-sdk/cerebras/@ai-sdk/provider-utils": ["@ai-sdk/provider-utils@4.0.33", "", { "dependencies": { "@ai-sdk/provider": "3.0.12", "@standard-schema/spec": "^1.1.0", "eventsource-parser": "^3.0.8" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-nJ0bAfegMAIJtrzMJtbzer1cS3nb7c7DsyU1S4nrPm7ZU0Mn6SBBZv5IGZZGTbpWTJwqKTSPeZJTXalbAxt1BA=="], + "opencode/@ai-sdk/openai/@ai-sdk/provider": ["@ai-sdk/provider@3.0.14", "", { "dependencies": { "json-schema": "^0.4.0" } }, "sha512-5X1k57JBJ4H7H1QjX7CnJYAB1I19r/trVZTMcSms7/kLNZ8RaU4Nt2agcwZzv82Hfx6Q7/TOLU7agAKeFfc8cA=="], + + "opencode/@ai-sdk/openai/@ai-sdk/provider-utils": ["@ai-sdk/provider-utils@4.0.38", "", { "dependencies": { "@ai-sdk/provider": "3.0.14", "@standard-schema/spec": "^1.1.0", "eventsource-parser": "^3.0.8" }, "peerDependencies": { "zod": "^3.25.76 || ^4.1.8" } }, "sha512-/HHGmtKllqjg1OLc023v9w9kK3laW7Z6TzfZukYQWCsGBbzB9p60zTvvpXFVcs44NZBVXL3viOa1HRKUbeee8g=="], + "p-locate/p-limit/yocto-queue": ["yocto-queue@0.1.0", "", {}, "sha512-rVksvsnNCdJ/ohGc6xgPwyN8eheCxsiLM8mxuE/t/mOVqJewPuO1miLpTHQiRgTKCLexL4MeAFVagts7HmNZ2Q=="], "pkg-dir/find-up/locate-path": ["locate-path@5.0.0", "", { "dependencies": { "p-locate": "^4.1.0" } }, "sha512-t7hw9pI+WvuwNJXwk5zVHpyhIqzg2qTlklJOf0mVxGSbe3Fp2VieZcduNYjaLDoy6p9uGpQEGWG87WpMKlNq8g=="], @@ -7369,6 +7377,8 @@ "@octokit/rest/@octokit/core/@octokit/types/@octokit/openapi-types": ["@octokit/openapi-types@27.0.0", "", {}, "sha512-whrdktVs1h6gtR+09+QsNk2+FO+49j6ga1c55YZudfEG+oKJVvJLQi3zkOm5JjiUXAagWK2tI2kTGKJ2Ys7MGA=="], + "@opencode-ai/core/@ai-sdk/openai/@ai-sdk/provider-utils/@standard-schema/spec": ["@standard-schema/spec@1.1.0", "", {}, "sha512-l2aFy5jALhniG5HgqrD6jXLi/rUWrKvqN/qJx6yoJsgKhblVd+iqqU4RCXavm/jPityDo5TCvKMnpjKnOriy0w=="], + "@opencode-ai/desktop/@actions/artifact/@actions/http-client/undici": ["undici@5.29.0", "", { "dependencies": { "@fastify/busboy": "^2.0.0" } }, "sha512-raqeBD6NQK4SkWhQzeYKd1KmIG6dllBOTt55Rmkt4HtI9mwdWtJljnrXjAFUBLTSN67HWrOIZ3EPF4kjUw80Bg=="], "@sentry/bundler-plugin-core/glob/minimatch/brace-expansion": ["brace-expansion@2.1.1", "", { "dependencies": { "balanced-match": "^1.0.0" } }, "sha512-WR1cURNjuvBLMZBMbqM0UoE+WAfdUcEV1ccD8PVBVOI+Z3ND4+SZbN8RsfT2bMuG1qwz5RFvPukSZm5fF2D5eA=="], @@ -7455,6 +7465,8 @@ "opencode/@ai-sdk/cerebras/@ai-sdk/provider-utils/@standard-schema/spec": ["@standard-schema/spec@1.1.0", "", {}, "sha512-l2aFy5jALhniG5HgqrD6jXLi/rUWrKvqN/qJx6yoJsgKhblVd+iqqU4RCXavm/jPityDo5TCvKMnpjKnOriy0w=="], + "opencode/@ai-sdk/openai/@ai-sdk/provider-utils/@standard-schema/spec": ["@standard-schema/spec@1.1.0", "", {}, "sha512-l2aFy5jALhniG5HgqrD6jXLi/rUWrKvqN/qJx6yoJsgKhblVd+iqqU4RCXavm/jPityDo5TCvKMnpjKnOriy0w=="], + "pkg-dir/find-up/locate-path/p-locate": ["p-locate@4.1.0", "", { "dependencies": { "p-limit": "^2.2.0" } }, "sha512-R79ZZ/0wAxKGu3oYMlz8jy/kbhsNrS7SKZ7PxEHBgJ5+F2mtFW2fK2cOtBh1cHYkQsbzFV7I+EoRKe6Yt0oK7A=="], "pkg-up/find-up/locate-path/p-locate": ["p-locate@3.0.0", "", { "dependencies": { "p-limit": "^2.0.0" } }, "sha512-x+12w/To+4GFfgJhBEpiDcLozRJGegY+Ei7/z0tSLkMmxGZNybVMSfWj9aJn8Z5Fc7dBUNJOOVgPv2H7IwulSQ=="], diff --git a/infra/lake.ts b/infra/lake.ts index dd11ca38f977..d4e46168e366 100644 --- a/infra/lake.ts +++ b/infra/lake.ts @@ -67,6 +67,10 @@ const athenaWorkgroup = new aws.athena.Workgroup("LakeAthenaWorkgroup", { configuration: { enforceWorkgroupConfiguration: true, publishCloudwatchMetricsEnabled: true, + // Athena bills $5/TB scanned; kill any query that would scan more than 2 TB + // so a regression cannot silently burn money. Stats sync full passes scan + // ~250 GB as of 2026-07. + bytesScannedCutoffPerQuery: 2 * 1024 ** 4, resultConfiguration: { outputLocation: $interpolate`s3://${athenaResultsBucket.bucket}/`, }, diff --git a/infra/stats.ts b/infra/stats.ts index b5b0e1c600e5..10d37119f0d7 100644 --- a/infra/stats.ts +++ b/infra/stats.ts @@ -185,7 +185,9 @@ export const statSync = new sst.aws.Service("StatsSyncService", { cluster: lakeCluster, architecture: "arm64", cpu: "0.25 vCPU", - memory: "0.5 GB", + // 0.5 GB caused an OOM crash loop: every restart immediately re-ran the 4 Athena + // stats queries (~$5/pass) every ~5 minutes instead of hourly. + memory: "2 GB", image: { context: ".", dockerfile: "packages/stats/server/Dockerfile", diff --git a/nix/hashes.json b/nix/hashes.json index 50208fdbbeb7..d3d3e556fad3 100644 --- a/nix/hashes.json +++ b/nix/hashes.json @@ -1,8 +1,8 @@ { "nodeModules": { - "x86_64-linux": "sha256-JTtn+wXTXg+yklvIMDLcGFaYhTU6ZrCgKT9JTNEQ3gA=", - "aarch64-linux": "sha256-gXU6zyhvAZrZirkL/PlHdkHtEof/7PVSPCaE34Jnd4U=", - "aarch64-darwin": "sha256-Q0oTG3uzOlD/X2kJingLle529lKFoTpyCW2rHXOZ6iE=", - "x86_64-darwin": "sha256-LINvKHxPibTlJeNzfACQx0x+Yj5oROT6Du3I5AtqqXk=" + "x86_64-linux": "sha256-QlwoEmN1jiw9TqQgOutPOv4mEMCNUR0Mm3ivYIWlc1o=", + "aarch64-linux": "sha256-KS8unXRXev29zbPskvp2ddtfAKXJN8VfInXnmPqypBI=", + "aarch64-darwin": "sha256-T3tgbuodPitCx2U5JJIQFLkRguAFeiMM7flTsF2H5Iw=", + "x86_64-darwin": "sha256-aITMNkQpFYAChDz5jZCchFkSylsDhJJQoyPu9k23Hl8=" } } diff --git a/packages/app/e2e/regression/file-browser-sidebar-tab-switch.spec.ts b/packages/app/e2e/regression/file-browser-sidebar-tab-switch.spec.ts new file mode 100644 index 000000000000..136232c2114b --- /dev/null +++ b/packages/app/e2e/regression/file-browser-sidebar-tab-switch.spec.ts @@ -0,0 +1,150 @@ +import { base64Encode } from "@opencode-ai/core/util/encode" +import { expect, test, type Page } from "@playwright/test" +import { mockOpenCodeServer } from "../utils/mock-server" +import { expectSessionTitle } from "../utils/waits" + +const directory = "C:/OpenCode/FileBrowserSidebar" +const projectID = "proj_file_browser_sidebar" +const sessionID = "ses_file_browser_sidebar" +const title = "File browser sidebar" +const server = `http://${process.env.PLAYWRIGHT_SERVER_HOST ?? "127.0.0.1"}:${process.env.PLAYWRIGHT_SERVER_PORT ?? "4096"}` +const files = Array.from({ length: 80 }, (_, index) => `file-${String(index).padStart(2, "0")}.ts`) +// Marks the file-browser sidebar DOM node so a remount (fresh node) is detectable. +const PROBE = "original" + +test.use({ viewport: { width: 1440, height: 900 } }) + +// The file-browser sidebar must stay mounted across preview/pinned file-tab +// switches. Remounting resets scroll and filter state. +test("keeps the file-browser sidebar mounted when switching file tabs", async ({ page }) => { + await setup(page) + + await page.goto(`/server/${base64Encode(server)}/session/${sessionID}`) + await expectSessionTitle(page, title) + + const panel = page.locator("#review-panel") + await panel.getByRole("button", { name: "Open file" }).click() + await expect(panel.getByRole("tab", { name: "Open file" })).toHaveAttribute("data-selected", "") + + const sidebar = panel.locator('[data-component="session-review-v2-sidebar-root"]') + await expect(sidebar).toBeVisible() + await expect(panel.getByRole("button", { name: "file-00.ts" })).toBeVisible() + + await panel.getByRole("button", { name: "file-00.ts" }).click() + await expect(panel.getByRole("tab", { name: "file-00.ts" })).toHaveAttribute("data-selected", "") + await expect(panel.getByText("contents:file-00.ts", { exact: true })).toBeVisible() + + const viewport = panel.locator('[data-slot="session-review-v2-sidebar-tree"] .scroll-view__viewport') + await viewport.hover() + await page.mouse.wheel(0, 100_000) + await expect + .poll(() => viewport.evaluate((element) => element.scrollHeight - element.clientHeight - element.scrollTop)) + .toBeLessThanOrEqual(1) + const scrolled = await viewport.evaluate((element) => element.scrollTop) + expect(scrolled).toBeGreaterThan(0) + await writeProbe(page) + + await panel.getByRole("button", { name: "file-79.ts" }).click() + await expect(panel.getByRole("tab", { name: "file-79.ts" })).toHaveAttribute("data-selected", "") + await expect(panel.getByText("contents:file-79.ts", { exact: true })).toBeVisible() + expect(await readProbe(page)).toBe(PROBE) + await expect.poll(() => viewport.evaluate((element) => element.scrollTop)).toBe(scrolled) + + await panel.getByRole("button", { name: "file-78.ts" }).dblclick() + await expect(panel.getByRole("tab", { name: "file-78.ts" })).toHaveAttribute("data-selected", "") + await panel.getByRole("button", { name: "file-79.ts" }).click() + await expect(panel.getByRole("tab", { name: "file-79.ts" })).toHaveAttribute("data-selected", "") + await panel.getByRole("tab", { name: "file-78.ts" }).click() + await expect(panel.getByRole("tab", { name: "file-78.ts" })).toHaveAttribute("data-selected", "") + expect(await readProbe(page)).toBe(PROBE) + await expect.poll(() => viewport.evaluate((element) => element.scrollTop)).toBe(scrolled) +}) + +type Probed = HTMLElement & { __e2eProbe?: string } + +async function writeProbe(page: Page) { + await page.locator('#review-panel [data-component="session-review-v2-sidebar-root"]').evaluate((el, probe) => { + ;(el as Probed).__e2eProbe = probe + }, PROBE) +} + +async function readProbe(page: Page) { + return page + .locator('#review-panel [data-component="session-review-v2-sidebar-root"]') + .evaluate((el) => (el as Probed).__e2eProbe) +} + +async function setup(page: Page) { + await mockOpenCodeServer(page, { + directory, + project: { + id: projectID, + worktree: directory, + vcs: "git", + name: "file-browser-sidebar", + time: { created: 1700000000000, updated: 1700000000000 }, + sandboxes: [], + }, + provider: { + all: [ + { + id: "opencode", + name: "OpenCode", + models: { test: { id: "test", name: "Test", limit: { context: 200_000 } } }, + }, + ], + connected: ["opencode"], + default: { providerID: "opencode", modelID: "test" }, + }, + sessions: [ + { + id: sessionID, + slug: sessionID, + projectID, + directory, + title, + version: "dev", + time: { created: 1700000000000, updated: 1700000000000 }, + }, + ], + vcsDiff: [], + fileList: (path) => { + if (path) return [] + return files.map((name) => ({ + name, + path: name, + absolute: `${directory}/${name}`, + type: "file" as const, + ignored: false, + })) + }, + fileContent: (path) => ({ type: "text", content: `contents:${path}` }), + pageMessages: () => ({ items: [] }), + }) + + await page.addInitScript( + ({ directory, server, sessionID }) => { + localStorage.setItem("settings.v3", JSON.stringify({ general: { newLayoutDesigns: true } })) + localStorage.setItem( + "opencode.global.dat:server", + JSON.stringify({ + projects: { local: [{ worktree: directory, expanded: true }] }, + lastProject: { local: directory }, + }), + ) + localStorage.setItem( + "opencode.global.dat:layout", + JSON.stringify({ review: { diffStyle: "split", panelOpened: true } }), + ) + localStorage.setItem( + "opencode.global.dat:review-panel-v2", + JSON.stringify({ sidebarOpened: true, sidebarWidth: 240, expandMode: "collapse" }), + ) + localStorage.setItem( + "opencode.window.browser.dat:tabs", + JSON.stringify([{ type: "session", server, sessionId: sessionID }]), + ) + }, + { directory, server, sessionID }, + ) +} diff --git a/packages/app/e2e/regression/review-open-file.spec.ts b/packages/app/e2e/regression/review-open-file.spec.ts index f69591f608a6..25ebd3a37ae6 100644 --- a/packages/app/e2e/regression/review-open-file.spec.ts +++ b/packages/app/e2e/regression/review-open-file.spec.ts @@ -74,6 +74,10 @@ test("opens and searches project files inline", async ({ page }) => { "opencode.global.dat:layout", JSON.stringify({ review: { diffStyle: "split", panelOpened: true } }), ) + localStorage.setItem( + "opencode.global.dat:review-panel-v2", + JSON.stringify({ sidebarOpened: false, sidebarWidth: 240, expandMode: "collapse" }), + ) localStorage.setItem( "opencode.window.browser.dat:tabs", JSON.stringify([{ type: "session", server, sessionId: sessionID }]), @@ -86,13 +90,18 @@ test("opens and searches project files inline", async ({ page }) => { await expectSessionTitle(page, title) const panel = page.locator("#review-panel") + const sidebar = panel.locator('[data-slot="session-review-v2-sidebar"]') + const sidebarToggle = panel.getByRole("button", { name: "Toggle file tree" }) const contextButton = page.getByRole("button", { name: "View context usage" }) await contextButton.click() await expect(panel.getByRole("tab", { name: "Context" })).toHaveAttribute("data-selected", "") await panel.getByRole("button", { name: "Open file" }).click() await expect(panel.getByRole("tab", { name: "Open file" })).toHaveAttribute("data-selected", "") + await expect(sidebarToggle).toBeDisabled() + await expect(sidebar).toBeVisible() await contextButton.click() await expect(panel.getByRole("tab", { name: "Context" })).toHaveAttribute("data-selected", "") + await expect(sidebar).toBeHidden() await panel.getByRole("button", { name: "Open file" }).click() const filter = panel.getByRole("combobox", { name: "Filter files" }) await expect(filter).toBeFocused() @@ -101,10 +110,13 @@ test("opens and searches project files inline", async ({ page }) => { await panel.getByRole("button", { name: "README.md" }).click() await expect(panel.getByRole("tab", { name: "README.md" })).toHaveAttribute("data-selected", "") + await expect(sidebarToggle).toBeEnabled() await expect(panel.getByText("contents:README.md", { exact: true })).toBeVisible() + await expect(sidebar).toHaveCount(0) await panel.getByRole("button", { name: "Open file" }).click() await expect(panel.getByRole("tab", { name: "README.md" })).toHaveCount(0) + await expect(sidebar).toBeVisible() await filter.fill("nested") const result = panel.getByRole("option", { name: /nested\.ts/ }) await expect(result).toBeVisible() @@ -113,12 +125,17 @@ test("opens and searches project files inline", async ({ page }) => { await expect(filter).toHaveAttribute("aria-activedescendant", resultID!) await filter.press("Enter") await expect(panel.getByRole("tab", { name: "nested.ts" })).toHaveAttribute("data-selected", "") + await expect(sidebarToggle).toBeEnabled() await expect(panel.getByText("contents:src/nested.ts", { exact: true })).toBeVisible() expect(searches).toContainEqual({ query: "nested", dirs: "false", limit: 200 }) await panel.getByRole("button", { name: "Open file" }).click() await expect(panel.getByRole("tab", { name: "nested.ts" })).toHaveCount(1) await expect(panel.getByRole("tab", { name: "Open file" })).toHaveAttribute("data-selected", "") + await expect(sidebarToggle).toBeDisabled() + await panel.getByRole("tab", { name: /Review/ }).click() + await expect(sidebarToggle).toBeEnabled() + await panel.getByRole("tab", { name: "Open file" }).click() await page.keyboard.press("Control+w") await expect(panel.getByRole("tab", { name: "Open file" })).toHaveCount(0) await expect(panel.getByRole("tab", { name: "nested.ts" })).toHaveAttribute("data-selected", "") diff --git a/packages/app/e2e/regression/review-tab-switch.spec.ts b/packages/app/e2e/regression/review-tab-switch.spec.ts index c2ea406c5ab3..8634166e5128 100644 --- a/packages/app/e2e/regression/review-tab-switch.spec.ts +++ b/packages/app/e2e/regression/review-tab-switch.spec.ts @@ -28,8 +28,8 @@ test("keeps the v2 review pane mounted when switching session tabs in a workspac await expectSessionTitle(page, titleA) await page.getByRole("button", { name: "Toggle review" }).click() - const reviewTab = page.getByRole("tab", { name: /Review/ }) - const reviewTabPanel = page.getByRole("tabpanel", { name: /Review/ }) + const reviewTab = page.locator("#session-side-panel-review-tab") + const reviewTabPanel = page.locator("#session-side-panel-review-tabpanel") await expect(reviewTab).toHaveAttribute("aria-controls", "session-side-panel-review-tabpanel") await expect(reviewTabPanel).toHaveAttribute("id", "session-side-panel-review-tabpanel") const review = page.locator('#review-panel [data-component="session-review-v2"]') diff --git a/packages/app/e2e/regression/review-terminal-stacked.spec.ts b/packages/app/e2e/regression/review-terminal-stacked.spec.ts index e3ba607b3d84..154bab48c47f 100644 --- a/packages/app/e2e/regression/review-terminal-stacked.spec.ts +++ b/packages/app/e2e/regression/review-terminal-stacked.spec.ts @@ -111,7 +111,7 @@ test("keeps the review tree and terminal sized when both panels are open", async await expectTree(page, 8, "git-0.ts") await selectMode(page, "Git changes", "Branch changes") - await expect(page.getByRole("tab", { name: "Review 2740" })).toBeVisible() + await expect(page.locator("#session-side-panel-review-tab")).toHaveText("Files Changed 2740") await page.keyboard.press("Control+Backquote") await expect(page.locator("#terminal-panel")).toBeVisible() await expectTree(page, 2_773, "action.yml") diff --git a/packages/app/e2e/regression/session-timeline-projection.spec.ts b/packages/app/e2e/regression/session-timeline-projection.spec.ts index 9fd2ca8d0b63..b1aabcc32ca9 100644 --- a/packages/app/e2e/regression/session-timeline-projection.spec.ts +++ b/packages/app/e2e/regression/session-timeline-projection.spec.ts @@ -152,7 +152,10 @@ test.describe("session timeline projection", () => { parentID: "msg_2000_diff_next_user", created: 1700000011000, }) - await setupTimeline(page, { messages: [user, assistantMessage(), nextUser, nextAssistant] }) + await setupTimeline(page, { + messages: [user, assistantMessage(), nextUser, nextAssistant], + settings: { newLayoutDesigns: false }, + }) const scroller = page.locator(".scroll-view__viewport", { has: page.locator("[data-timeline-row]") }) await scroller.evaluate((element) => (element.scrollTop = 0)) diff --git a/packages/app/e2e/regression/session-timeline-shell-outline.spec.ts b/packages/app/e2e/regression/session-timeline-shell-outline.spec.ts new file mode 100644 index 000000000000..54139cc3711b --- /dev/null +++ b/packages/app/e2e/regression/session-timeline-shell-outline.spec.ts @@ -0,0 +1,228 @@ +import { expect, test, type Locator, type Page } from "@playwright/test" +import { + assistantMessage, + setupTimeline, + shell, + textPart, + toolPart, + userMessage, +} from "../performance/timeline-stability/fixture" + +for (const deviceScaleFactor of [1.25, 1.5]) { + test(`keeps the shell outline inside a fractionally short virtual row at ${deviceScaleFactor}x`, async ({ page }) => { + const shellID = "prt_shell_outline" + const timeline = await setupTimeline(page, { + messages: [userMessage(), assistantMessage([shell(shellID, "completed", "shell output")])], + settings: { newLayoutDesigns: true, shellToolPartsExpanded: true }, + reducedMotion: true, + deviceScaleFactor, + }) + const part = page.locator(`[data-timeline-part-id="${shellID}"]`) + const output = part.locator('[data-component="bash-output"]') + const row = page.locator("[data-timeline-key]", { has: part }) + await expect(output).toBeVisible() + await timeline.settle() + + const geometry = await row.evaluate((element) => { + const output = element.querySelector('[data-component="bash-output"]') + if (!output) throw new Error("Shell output is unavailable") + const rowRect = element.getBoundingClientRect() + const outputRect = output.getBoundingClientRect() + // Match a rounded-down measurement at a fractional device-pixel phase. + element.style.height = `${outputRect.bottom - rowRect.top - 0.49}px` + element.style.transform = "translateY(0.25px)" + output.style.setProperty("--v2-border-border-base", "rgb(255, 0, 255)") + output.style.setProperty("background", "rgb(0, 0, 0)", "important") + const style = getComputedStyle(output) + return { + outputWidth: outputRect.width, + outputHeight: outputRect.height, + borderColor: style.borderTopColor, + boxShadow: style.boxShadow, + clipMargin: getComputedStyle(element).overflowClipMargin, + } + }) + await timeline.settle() + + const clipped = await row.evaluate((element) => { + const output = element.querySelector('[data-component="bash-output"]')! + return output.getBoundingClientRect().bottom - element.getBoundingClientRect().bottom + }) + expect(clipped).toBeCloseTo(0.49, 1) + + expect(await page.evaluate(() => devicePixelRatio)).toBe(deviceScaleFactor) + const edges = await captureCardEdges(page, output) + + expect(edges.box.width).toBeCloseTo(geometry.outputWidth, 2) + expect(edges.box.height).toBeCloseTo(geometry.outputHeight, 2) + expect(geometry.borderColor).toBe("rgb(255, 0, 255)") + expect(geometry.boxShadow).toBe("none") + expect(geometry.clipMargin).toBe("0.5px") + expect(edges.magenta.top).toBeGreaterThan(0.75) + expect(edges.magenta.bottom).toBeGreaterThan(0.75) + expect(edges.magenta.vertical).toBeGreaterThanOrEqual(2) + }) +} + +test("keeps the patch card inside a fractionally short virtual row", async ({ page }) => { + const patchID = "prt_patch_outline" + const file = { + filePath: "src/outline.ts", + relativePath: "src/outline.ts", + type: "update", + additions: 1, + deletions: 1, + before: "const outline = false\n", + after: "const outline = true\n", + } + const timeline = await setupTimeline(page, { + messages: [ + userMessage(), + assistantMessage([ + toolPart(patchID, "apply_patch", "completed", { files: [file.filePath] }, { metadata: { files: [file] } }), + ]), + ], + settings: { editToolPartsExpanded: true, newLayoutDesigns: true }, + reducedMotion: true, + }) + const part = page.locator(`[data-timeline-part-id="${patchID}"]`) + const card = part.locator('[data-component="accordion"][data-scope="apply-patch"]') + const row = page.locator("[data-timeline-key]", { has: part }) + await expect(card).toBeVisible() + await timeline.settle() + + const geometry = await row.evaluate((element) => { + const card = element.querySelector('[data-component="accordion"][data-scope="apply-patch"]') + if (!card) throw new Error("Patch card is unavailable") + const rowRect = element.getBoundingClientRect() + const cardRect = card.getBoundingClientRect() + element.style.height = `${cardRect.bottom - rowRect.top - 0.49}px` + const clipMargin = getComputedStyle(element).overflowClipMargin + const bottom = element.getBoundingClientRect().bottom + return { + overflow: card.getBoundingClientRect().bottom - bottom, + paintOverflow: card.getBoundingClientRect().bottom - bottom - Number.parseFloat(clipMargin), + clipMargin, + cardWidth: cardRect.width, + cardHeight: cardRect.height, + } + }) + await timeline.settle() + + expect(geometry.overflow).toBeCloseTo(0.49, 1) + expect(geometry.paintOverflow).toBeLessThanOrEqual(0) + const edges = await captureCardEdges(page, card) + expect(edges.box.width).toBeCloseTo(geometry.cardWidth, 2) + expect(edges.box.height).toBeCloseTo(geometry.cardHeight, 2) + expect(edges.luminance.top).toBeLessThan(245) + expect(edges.luminance.bottom).toBeLessThan(245) + expect(Math.abs(edges.luminance.bottom - edges.luminance.top)).toBeLessThan(10) + expect(geometry.clipMargin).toBe("0.5px") +}) + +test("allows paint rounding for every framed row but not fixed turn gaps", async ({ page }) => { + const secondUserID = "msg_outline_second_user" + await setupTimeline(page, { + messages: [ + userMessage(undefined, { + summary: { + diffs: [ + { + file: "src/summary.ts", + additions: 1, + deletions: 1, + patch: "@@ -1 +1 @@\n-export const value = 1\n+export const value = 2", + }, + ], + }, + }), + assistantMessage([textPart("prt_outline_text", "Assistant text")]), + userMessage(undefined, { id: secondUserID, created: 1700000010000 }), + assistantMessage([], { + id: "msg_outline_second_assistant", + parentID: secondUserID, + created: 1700000011000, + }), + ], + }) + await expect(page.locator('[data-timeline-row="DiffSummary"]')).toBeVisible() + await expect(page.locator('[data-timeline-row="TurnGap"]')).toBeVisible() + + const rows = await page.locator("[data-timeline-key]").evaluateAll((elements) => + elements.map((element) => ({ + tag: element.querySelector("[data-timeline-row]")?.dataset.timelineRow, + clipMargin: getComputedStyle(element).overflowClipMargin, + })), + ) + expect(rows.filter((row) => row.tag !== "TurnGap").every((row) => row.clipMargin === "0.5px")).toBe(true) + expect(rows.filter((row) => row.tag === "TurnGap")).toEqual([{ tag: "TurnGap", clipMargin: "0px" }]) +}) + +async function captureCardEdges(page: Page, card: Locator) { + const box = await card.boundingBox() + if (!box) throw new Error("Tool card bounds are unavailable") + const viewport = page.viewportSize() + if (!viewport) throw new Error("Viewport bounds are unavailable") + const screenshot = await page.screenshot() + return page.evaluate( + async ({ source, box, viewport }) => { + const image = new Image() + image.src = source + await image.decode() + const canvas = document.createElement("canvas") + canvas.width = image.naturalWidth + canvas.height = image.naturalHeight + const context = canvas.getContext("2d") + if (!context) throw new Error("2D canvas is unavailable") + context.drawImage(image, 0, 0) + const scale = { + x: image.naturalWidth / viewport.width, + y: image.naturalHeight / viewport.height, + } + const rows = (candidates: number[]) => { + const left = Math.floor((box.x + 8) * scale.x) + const width = Math.floor((box.width - 16) * scale.x) + return candidates.map((row) => { + const pixels = context.getImageData(left, row, width, 1).data + const indexes = Array.from({ length: width }, (_, index) => index * 4) + return { + luminance: + indexes + .map((index) => (pixels[index]! + pixels[index + 1]! + pixels[index + 2]!) / 3) + .reduce((sum, value) => sum + value, 0) / width, + magenta: + indexes.filter((index) => pixels[index]! > 200 && pixels[index + 1]! < 180 && pixels[index + 2]! > 200) + .length / width, + } + }) + } + const pixels = context.getImageData(0, 0, image.naturalWidth, image.naturalHeight).data + const columns = new Uint32Array(image.naturalWidth) + for (let index = 0; index < pixels.length; index += 4) { + if (pixels[index]! <= 200 || pixels[index + 1]! >= 180 || pixels[index + 2]! <= 200) continue + columns[(index / 4) % image.naturalWidth] = columns[(index / 4) % image.naturalWidth]! + 1 + } + const top = box.y * scale.y + const bottom = (box.y + box.height) * scale.y + const topRows = rows([Math.floor(top) - 1, Math.floor(top), Math.ceil(top)]) + const bottomRows = rows([Math.floor(bottom) - 2, Math.floor(bottom) - 1, Math.ceil(bottom) - 1]) + return { + box, + luminance: { + top: Math.min(...topRows.map((row) => row.luminance)), + bottom: rows([Math.ceil(bottom) - 1])[0]!.luminance, + }, + magenta: { + top: Math.max(...topRows.map((row) => row.magenta)), + bottom: Math.max(...bottomRows.map((row) => row.magenta)), + vertical: Array.from(columns).filter((count) => count > box.height * scale.y * 0.75).length, + }, + } + }, + { + source: `data:image/png;base64,${screenshot.toString("base64")}`, + viewport, + box, + }, + ) +} diff --git a/packages/app/package.json b/packages/app/package.json index 510e2015dbe5..c5d9a111985a 100644 --- a/packages/app/package.json +++ b/packages/app/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/app", - "version": "1.17.18", + "version": "1.17.19", "description": "", "type": "module", "exports": { diff --git a/packages/app/src/components/dialog-edit-project-v2.tsx b/packages/app/src/components/dialog-edit-project-v2.tsx new file mode 100644 index 000000000000..dc9714aa0625 --- /dev/null +++ b/packages/app/src/components/dialog-edit-project-v2.tsx @@ -0,0 +1,156 @@ +import { ButtonV2 } from "@opencode-ai/ui/v2/button-v2" +import { Dialog, DialogBody, DialogFooter, DialogHeader, DialogTitle } from "@opencode-ai/ui/v2/dialog-v2" +import { DividerV2 } from "@opencode-ai/ui/v2/divider-v2" +import { Field } from "@opencode-ai/ui/v2/field-v2" +import { Icon } from "@opencode-ai/ui/v2/icon" +import { ProjectAvatar, PROJECT_AVATAR_VARIANTS } from "@opencode-ai/ui/v2/project-avatar-v2" +import { TextareaV2 } from "@opencode-ai/ui/v2/textarea-v2" +import { TextInputV2 } from "@opencode-ai/ui/v2/text-input-v2" +import { For, Show } from "solid-js" +import { useLanguage } from "@/context/language" +import { getProjectAvatarVariant, type LocalProject } from "@/context/layout" +import { ServerConnection } from "@/context/server" +import { getProjectAvatarSource } from "@/pages/layout/helpers" +import { createEditProjectModel } from "./edit-project" + +export function DialogEditProjectV2(props: { project: LocalProject; server: ServerConnection.Any }) { + const language = useLanguage() + const model = createEditProjectModel(props) + + return ( + +
+ + {language.t("dialog.project.edit.title")} + + + + + {language.t("dialog.project.edit.name")} + model.setStore("name", event.currentTarget.value)} + /> + + +
+
+ {language.t("dialog.project.edit.icon")} +
+
+ + { + model.setIconInput(element) + }} + type="file" + accept="image/*" + class="hidden" + onChange={model.inputChange} + /> +
+ {language.t("dialog.project.edit.icon.hint")} + {language.t("dialog.project.edit.icon.recommended")} +
+
+
+ + +
+
+ {language.t("dialog.project.edit.color")} +
+
+ + {(color) => ( + + )} + +
+
+
+ + + {language.t("dialog.project.edit.worktree.startup")} + {language.t("dialog.project.edit.worktree.startup.description")} + model.setStore("startup", event.currentTarget.value)} + /> + +
+ + + {language.t("common.cancel")} + + + {model.save.isPending ? language.t("common.saving") : language.t("common.save")} + + + +
+ ) +} diff --git a/packages/app/src/components/dialog-edit-project.tsx b/packages/app/src/components/dialog-edit-project.tsx index b861492af5ea..86a9630359ba 100644 --- a/packages/app/src/components/dialog-edit-project.tsx +++ b/packages/app/src/components/dialog-edit-project.tsx @@ -1,123 +1,32 @@ import { Button } from "@opencode-ai/ui/button" -import { useDialog } from "@opencode-ai/ui/context/dialog" import { Dialog } from "@opencode-ai/ui/dialog" import { TextField } from "@opencode-ai/ui/text-field" -import { useMutation } from "@tanstack/solid-query" import { Icon } from "@opencode-ai/ui/icon" -import { createMemo, For, Show } from "solid-js" -import { createStore } from "solid-js/store" +import { For, Show } from "solid-js" import { type LocalProject, getAvatarColors } from "@/context/layout" -import { getFilename } from "@opencode-ai/core/util/path" import { Avatar } from "@opencode-ai/ui/avatar" import { useLanguage } from "@/context/language" import { getProjectAvatarSource } from "@/pages/layout/helpers" import { ServerConnection } from "@/context/server" -import { useGlobal } from "@/context/global" +import { createEditProjectModel } from "./edit-project" const AVATAR_COLOR_KEYS = ["pink", "mint", "orange", "purple", "cyan", "lime"] as const export function DialogEditProject(props: { project: LocalProject; server: ServerConnection.Any }) { - const dialog = useDialog() - const global = useGlobal() const language = useLanguage() - const serverCtx = createMemo(() => global.ensureServerCtx(props.server)) - const serverSDK = () => serverCtx().sdk - const serverSync = () => serverCtx().sync - - const folderName = createMemo(() => getFilename(props.project.worktree)) - const defaultName = createMemo(() => props.project.name || folderName()) - - const [store, setStore] = createStore({ - name: defaultName(), - color: props.project.icon?.color, - iconOverride: props.project.icon?.override, - startup: props.project.commands?.start ?? "", - dragOver: false, - iconHover: false, - }) - - let iconInput: HTMLInputElement | undefined - - function handleFileSelect(file: File) { - if (!file.type.startsWith("image/")) return - const reader = new FileReader() - reader.onload = (e) => { - setStore("iconOverride", e.target?.result as string) - setStore("iconHover", false) - } - reader.readAsDataURL(file) - } - - function handleDrop(e: DragEvent) { - e.preventDefault() - setStore("dragOver", false) - const file = e.dataTransfer?.files[0] - if (file) handleFileSelect(file) - } - - function handleDragOver(e: DragEvent) { - e.preventDefault() - setStore("dragOver", true) - } - - function handleDragLeave() { - setStore("dragOver", false) - } - - function handleInputChange(e: Event) { - const input = e.target as HTMLInputElement - const file = input.files?.[0] - if (file) handleFileSelect(file) - } - - function clearIcon() { - setStore("iconOverride", "") - } - - const saveMutation = useMutation(() => ({ - mutationFn: async () => { - const name = store.name.trim() === folderName() ? "" : store.name.trim() - const start = store.startup.trim() - - if (props.project.id && props.project.id !== "global") { - await serverSDK().client.project.update({ - projectID: props.project.id, - directory: props.project.worktree, - name, - icon: { color: store.color || "", override: store.iconOverride || "" }, - commands: { start }, - }) - serverSync().project.icon(props.project.worktree, store.iconOverride || undefined) - dialog.close() - return - } - - serverSync().project.meta(props.project.worktree, { - name, - icon: { color: store.color || undefined, override: store.iconOverride || undefined }, - commands: { start: start || undefined }, - }) - dialog.close() - }, - })) - - function handleSubmit(e: SubmitEvent) { - e.preventDefault() - if (saveMutation.isPending) return - saveMutation.mutate() - } + const model = createEditProjectModel(props) return ( -
+
setStore("name", v)} + placeholder={model.folderName()} + value={model.store.name} + onChange={(v) => model.setStore("name", v)} />
@@ -125,38 +34,32 @@ export function DialogEditProject(props: { project: LocalProject; server: Server
setStore("iconHover", true)} - onMouseLeave={() => setStore("iconHover", false)} + onMouseEnter={() => model.setStore("iconHover", true)} + onMouseLeave={() => model.setStore("iconHover", false)} >
{ - if (store.iconOverride && store.iconHover) { - clearIcon() - } else { - iconInput?.click() - } + "border-text-interactive-base bg-surface-info-base/20": model.store.dragOver, + "border-border-base hover:border-border-strong": !model.store.dragOver, + "overflow-hidden": !!model.store.iconOverride, }} + onDrop={model.drop} + onDragOver={model.dragOver} + onDragLeave={model.dragLeave} + onClick={model.iconClick} >
@@ -174,8 +77,8 @@ export function DialogEditProject(props: { project: LocalProject; server: Server
@@ -183,8 +86,8 @@ export function DialogEditProject(props: { project: LocalProject; server: Server
@@ -193,12 +96,12 @@ export function DialogEditProject(props: { project: LocalProject; server: Server { - iconInput = el + model.setIconInput(el) }} type="file" accept="image/*" class="hidden" - onChange={handleInputChange} + onChange={model.inputChange} />
{language.t("dialog.project.edit.icon.hint")} @@ -207,7 +110,7 @@ export function DialogEditProject(props: { project: LocalProject; server: Server
- +
@@ -216,21 +119,21 @@ export function DialogEditProject(props: { project: LocalProject; server: Server
- -
diff --git a/packages/app/src/components/edit-project.ts b/packages/app/src/components/edit-project.ts new file mode 100644 index 000000000000..3ec999da0639 --- /dev/null +++ b/packages/app/src/components/edit-project.ts @@ -0,0 +1,119 @@ +import { getFilename } from "@opencode-ai/core/util/path" +import { useDialog } from "@opencode-ai/ui/context/dialog" +import { useMutation } from "@tanstack/solid-query" +import { createMemo } from "solid-js" +import { createStore } from "solid-js/store" +import { useGlobal } from "@/context/global" +import { type LocalProject } from "@/context/layout" +import { ServerConnection } from "@/context/server" + +export function createEditProjectModel(props: { project: LocalProject; server: ServerConnection.Any }) { + const dialog = useDialog() + const global = useGlobal() + const serverCtx = createMemo(() => global.ensureServerCtx(props.server)) + const folderName = createMemo(() => getFilename(props.project.worktree)) + const defaultName = createMemo(() => props.project.name || folderName()) + const [store, setStore] = createStore({ + name: defaultName(), + color: props.project.icon?.color, + iconOverride: props.project.icon?.override, + startup: props.project.commands?.start ?? "", + dragOver: false, + iconHover: false, + }) + let iconInput: HTMLInputElement | undefined + + function selectFile(file: File) { + if (!file.type.startsWith("image/")) return + const reader = new FileReader() + reader.onload = (event) => { + const result = event.target?.result + if (typeof result !== "string") return + setStore("iconOverride", result) + setStore("iconHover", false) + } + reader.readAsDataURL(file) + } + + function drop(event: DragEvent) { + event.preventDefault() + setStore("dragOver", false) + const file = event.dataTransfer?.files[0] + if (file) selectFile(file) + } + + function dragOver(event: DragEvent) { + event.preventDefault() + setStore("dragOver", true) + } + + function dragLeave() { + setStore("dragOver", false) + } + + function inputChange(event: Event) { + const file = (event.currentTarget as HTMLInputElement).files?.[0] + if (file) selectFile(file) + } + + function iconClick() { + if (store.iconOverride && store.iconHover) { + setStore("iconOverride", "") + return + } + iconInput?.click() + } + + const save = useMutation(() => ({ + mutationFn: async () => { + const name = store.name.trim() === folderName() ? "" : store.name.trim() + const start = store.startup.trim() + + if (props.project.id && props.project.id !== "global") { + await serverCtx().sdk.client.project.update({ + projectID: props.project.id, + directory: props.project.worktree, + name, + icon: { color: store.color || "", override: store.iconOverride || "" }, + commands: { start }, + }) + serverCtx().sync.project.icon(props.project.worktree, store.iconOverride || undefined) + dialog.close() + return + } + + serverCtx().sync.project.meta(props.project.worktree, { + name, + icon: { color: store.color || undefined, override: store.iconOverride || undefined }, + commands: { start: start || undefined }, + }) + dialog.close() + }, + })) + + function submit(event: SubmitEvent) { + event.preventDefault() + if (save.isPending) return + save.mutate() + } + + return { + store, + setStore, + folderName, + defaultName, + save, + submit, + drop, + dragOver, + dragLeave, + inputChange, + iconClick, + close() { + dialog.close() + }, + setIconInput(input: HTMLInputElement) { + iconInput = input + }, + } +} diff --git a/packages/app/src/components/file-tree-v2-model.test.ts b/packages/app/src/components/file-tree-v2-model.test.ts index 288f7112e929..f2a63fe86d3d 100644 --- a/packages/app/src/components/file-tree-v2-model.test.ts +++ b/packages/app/src/components/file-tree-v2-model.test.ts @@ -1,8 +1,9 @@ import { describe, expect, test } from "bun:test" -import { buildFileTreeV2Model, flattenFileTreeV2 } from "./file-tree-v2-model" +import { buildFileTreeV2Model, flattenFileTreeV2, flattenLiveFileTreeV2 } from "./file-tree-v2-model" +import type { FileNode } from "@opencode-ai/sdk/v2" -describe("file tree v2 model", () => { - test("builds sorted depth-first rows", () => { +describe("buildFileTreeV2Model", () => { + test("builds a sorted tree and flattens expanded directories", () => { const model = buildFileTreeV2Model(["src/z.ts", "src/lib/b.ts", "src/lib/a.ts", "README.md", "docs/guide.md"]) expect(model.total).toBe(8) @@ -18,7 +19,7 @@ describe("file tree v2 model", () => { ]) }) - test("omits descendants of collapsed directories", () => { + test("skips children of collapsed directories", () => { const model = buildFileTreeV2Model(["src/lib/a.ts", "src/z.ts"]) expect(flattenFileTreeV2(model, (path) => path !== "src/lib").map((row) => row.node.path)).toEqual([ @@ -28,19 +29,46 @@ describe("file tree v2 model", () => { ]) }) - test("normalizes separators and duplicate paths", () => { + test("normalizes duplicate and messy paths", () => { const model = buildFileTreeV2Model(["src\\lib\\a.ts", "src/lib/a.ts", "/src//lib/b.ts/"]) const rows = flattenFileTreeV2(model, () => true) - expect(model.total).toBe(4) expect(rows.map((row) => row.node.path)).toEqual(["src", "src/lib", "src/lib/a.ts", "src/lib/b.ts"]) expect(rows.find((row) => row.node.path === "src/lib/a.ts")?.node.originalPath).toBe("src\\lib\\a.ts") }) - test("supports paths deeper than the legacy recursion limit", () => { - const file = `${Array.from({ length: 130 }, (_, index) => `dir-${index}`).join("/")}/file.ts` + test("handles deeply nested paths", () => { + const file = Array.from({ length: 130 }, (_, index) => `d${index}`).join("/") + "/leaf.ts" const model = buildFileTreeV2Model([file]) expect(flattenFileTreeV2(model, () => true)).toHaveLength(131) }) }) + +describe("flattenLiveFileTreeV2", () => { + test("flattens live children using original paths for nested lookups", () => { + const nodes: Record = { + "": [ + { name: "src", path: "src", absolute: "/repo/src", type: "directory", ignored: false }, + { name: "README.md", path: "README.md", absolute: "/repo/README.md", type: "file", ignored: false }, + ], + src: [ + { name: "a.ts", path: "src/a.ts", absolute: "/repo/src/a.ts", type: "file", ignored: false }, + { name: "lib", path: "src/lib", absolute: "/repo/src/lib", type: "directory", ignored: false }, + ], + "src/lib": [{ name: "b.ts", path: "src/lib/b.ts", absolute: "/repo/src/lib/b.ts", type: "file", ignored: false }], + } + + expect( + flattenLiveFileTreeV2( + (path) => nodes[path] ?? [], + (path) => path === "src", + ).map((row) => [row.node.path, row.node.originalPath, row.level]), + ).toEqual([ + ["src", "src", 0], + ["src/a.ts", "src/a.ts", 1], + ["src/lib", "src/lib", 1], + ["README.md", "README.md", 0], + ]) + }) +}) diff --git a/packages/app/src/components/file-tree-v2-model.ts b/packages/app/src/components/file-tree-v2-model.ts index 2127b6800f86..27783a9bb16b 100644 --- a/packages/app/src/components/file-tree-v2-model.ts +++ b/packages/app/src/components/file-tree-v2-model.ts @@ -75,3 +75,33 @@ export function flattenFileTreeV2(model: FileTreeV2Model, expanded: (path: strin return rows } + +export function flattenLiveFileTreeV2( + children: (path: string) => readonly FileNode[], + expanded: (path: string) => boolean, +) { + const rows: FileTreeV2Row[] = [] + const stack = children("") + .toReversed() + .map((node) => ({ node: toLiveNode(node), level: 0 })) + + while (stack.length > 0) { + const row = stack.pop()! + rows.push(row) + if (row.node.type !== "directory" || !expanded(row.node.path)) continue + const nested = children(row.node.originalPath) + for (let index = nested.length - 1; index >= 0; index--) { + stack.push({ node: toLiveNode(nested[index]!), level: row.level + 1 }) + } + } + + return rows +} + +function toLiveNode(node: FileNode): FileTreeV2Node { + return { + ...node, + path: normalizeFileTreeV2Path(node.path), + originalPath: node.path, + } +} diff --git a/packages/app/src/components/file-tree-v2.tsx b/packages/app/src/components/file-tree-v2.tsx index c712a8044360..26cf49d411b8 100644 --- a/packages/app/src/components/file-tree-v2.tsx +++ b/packages/app/src/components/file-tree-v2.tsx @@ -16,7 +16,13 @@ import type { FileNode } from "@opencode-ai/sdk/v2" import { Icon } from "@opencode-ai/ui/v2/icon" import { pathToFileUrl, withFileDragImage, type Kind } from "@/components/file-tree" import { createVirtualizer, defaultRangeExtractor } from "@tanstack/solid-virtual" -import { buildFileTreeV2Model, flattenFileTreeV2, normalizeFileTreeV2Path } from "@/components/file-tree-v2-model" +import { + buildFileTreeV2Model, + flattenFileTreeV2, + flattenLiveFileTreeV2, + normalizeFileTreeV2Path, + type FileTreeV2Node, +} from "@/components/file-tree-v2-model" import { virtualScrollElement } from "@/components/virtual-scroll-element" export type { Kind } from "@/components/file-tree" @@ -36,7 +42,7 @@ function guideLineLeft(level: number) { export const kindLabel = (kind: Kind) => { if (kind === "add") return "A" if (kind === "del") return "D" - return "" + return "M" } export const kindChange = (kind: Kind) => { @@ -68,7 +74,7 @@ const FileTreeNodeV2 = ( "class", "classList", ]) - const kind = () => local.kinds?.get(local.node.path) + const kind = () => local.kinds?.get(normalizeFileTreeV2Path(local.node.path)) return ( - {(_, index) => ( -
- )} + {(_, index) =>
} ) } @@ -126,12 +127,18 @@ export default function FileTreeV2(props: { kinds?: ReadonlyMap draggable?: boolean onFileClick?: (file: FileNode) => void + onFileDoubleClick?: (file: FileNode) => void }) { const file = useFile() + const live = () => props.allowed === undefined const draggable = () => props.draggable ?? true const active = () => normalizeFileTreeV2Path(props.active ?? "") - const model = createMemo(() => buildFileTreeV2Model(props.allowed ?? [])) - const rows = createMemo(() => flattenFileTreeV2(model(), (path) => file.tree.state(path)?.expanded ?? true)) + const model = createMemo(() => (live() ? undefined : buildFileTreeV2Model(props.allowed ?? []))) + const expanded = (path: string) => file.tree.state(path)?.expanded ?? !live() + const rows = createMemo(() => { + if (live()) return flattenLiveFileTreeV2((path) => file.tree.children(path), expanded) + return flattenFileTreeV2(model()!, expanded) + }) const [root, setRoot] = createSignal() const [focused, setFocused] = createSignal() const virtualizer = createVirtualizer({ @@ -155,16 +162,49 @@ export default function FileTreeV2(props: { return [...indexes, index].sort((a, b) => a - b) }, }) + + createEffect(() => { + if (!live()) return + void file.tree.list("") + }) + + // Only scroll when the active path changes (or first appears in the tree). + // Do not re-scroll when expand/collapse reshuffles `rows()`. + let scrolledActive: string | undefined createEffect(() => { const path = active() - if (!path) return + if (!path) { + scrolledActive = undefined + return + } const index = rows().findIndex((row) => row.node.path === path) if (index < 0) return + if (scrolledActive === path) return + scrolledActive = path queueMicrotask(() => { - if (virtualizer.range && index >= virtualizer.range.startIndex && index <= virtualizer.range.endIndex) return - virtualizer.scrollToIndex(index, { align: "auto" }) + const next = rows().findIndex((row) => row.node.path === path) + if (next < 0) return + if (virtualizer.range && next >= virtualizer.range.startIndex && next <= virtualizer.range.endIndex) return + virtualizer.scrollToIndex(next, { align: "auto" }) }) }) + + const selectFile = (node: FileTreeV2Node, action?: (file: FileNode) => void) => { + action?.({ + ...node, + path: node.originalPath, + absolute: node.originalPath, + }) + } + + const toggleDirectory = (path: string, originalPath: string) => { + if (expanded(path)) { + file.tree.collapse(originalPath) + return + } + file.tree.expand(originalPath, live() ? undefined : { list: false }) + } + const rowByKey = createMemo(() => new Map(rows().map((row) => [row.node.path, row] as const))) const virtualItemByKey = createMemo( () => new Map(virtualizer.getVirtualItems().map((item) => [item.key, item] as const)), @@ -175,7 +215,7 @@ export default function FileTreeV2(props: {
@@ -209,13 +249,8 @@ export default function FileTreeV2(props: { class="relative" onFocus={() => setFocused(row().node.path)} onBlur={() => setFocused(undefined)} - onClick={() => - props.onFileClick?.({ - ...row().node, - path: row().node.originalPath, - absolute: row().node.originalPath, - }) - } + onClick={() => selectFile(row().node, props.onFileClick)} + onDblClick={() => selectFile(row().node, props.onFileDoubleClick)} > 0}> @@ -239,17 +274,13 @@ export default function FileTreeV2(props: { class="relative" onFocus={() => setFocused(row().node.path)} onBlur={() => setFocused(undefined)} - aria-expanded={file.tree.state(row().node.path)?.expanded ?? true} - onClick={() => - file.tree.state(row().node.path)?.expanded === false - ? file.tree.expand(row().node.path, { list: false }) - : file.tree.collapse(row().node.path) - } + aria-expanded={expanded(row().node.path)} + onClick={() => toggleDirectory(row().node.path, row().node.originalPath)} >
diff --git a/packages/app/src/components/prompt-input.tsx b/packages/app/src/components/prompt-input.tsx index 324a789825ae..7db0583c5167 100644 --- a/packages/app/src/components/prompt-input.tsx +++ b/packages/app/src/components/prompt-input.tsx @@ -19,6 +19,7 @@ import { selectionFromLines, type SelectedLineRange, useFile } from "@/context/f import { ContentPart, DEFAULT_PROMPT, + isCommentItem, isPromptEqual, Prompt, usePrompt, @@ -36,6 +37,8 @@ import { Icon } from "@opencode-ai/ui/icon" import { ProviderIcon } from "@opencode-ai/ui/provider-icon" import { Tooltip, TooltipKeybind } from "@opencode-ai/ui/tooltip" import { ButtonV2 } from "@opencode-ai/ui/v2/button-v2" +import { Icon as IconV2 } from "@opencode-ai/ui/v2/icon" +import { IconButtonV2 } from "@opencode-ai/ui/v2/icon-button-v2" import { KeybindV2 } from "@opencode-ai/ui/v2/keybind-v2" import { MenuV2 } from "@opencode-ai/ui/v2/menu-v2" import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" @@ -1333,6 +1336,7 @@ export const PromptInput: Component = (props) => { onQueue: props.onQueue, onAbort: props.onAbort, onSubmit: props.onSubmit, + model: props.controls.model.selection, }) const handleKeyDown = (event: KeyboardEvent) => { @@ -1623,7 +1627,7 @@ export const PromptInput: Component = (props) => { )} /> !isCommentItem(item))} active={(item) => { const active = comments.active() return !!item.commentID && item.commentID === active?.id && item.path === active?.file @@ -1633,6 +1637,7 @@ export const PromptInput: Component = (props) => { if (item.commentID) comments.remove(item.path, item.commentID) prompt.context.remove(item.key) }} + newLayoutDesigns={props.controls.newLayoutDesigns} t={(key) => language.t(key as Parameters[0])} /> = (props) => { } onRemove={removeAttachment} removeLabel={language.t("prompt.attachment.remove")} + newLayoutDesigns={props.controls.newLayoutDesigns} + comments={contextItems().filter(isCommentItem)} + commentActive={(item) => { + const active = comments.active() + return !!item.commentID && item.commentID === active?.id && item.path === active?.file + }} + onOpenComment={openComment} + onRemoveComment={(item) => { + if (item.commentID) comments.remove(item.path, item.commentID) + prompt.context.remove(item.key) + }} />
= (props) => { > } + variant="ghost-muted" + size="large" style={buttons()} disabled={store.mode !== "normal"} tabIndex={store.mode === "normal" ? undefined : -1} aria-label={language.t("prompt.menu.addImagesAndFiles")} /> - + {language.t("prompt.menu.imagesAndFiles")} @@ -1852,6 +1865,7 @@ export const PromptInput: Component = (props) => { if (item.commentID) comments.remove(item.path, item.commentID) prompt.context.remove(item.key) }} + newLayoutDesigns={props.controls.newLayoutDesigns} t={(key) => language.t(key as Parameters[0])} /> = (props) => { } onRemove={removeAttachment} removeLabel={language.t("prompt.attachment.remove")} + newLayoutDesigns={props.controls.newLayoutDesigns} />
)} - {props.state.modelName} + {props.state.modelName} diff --git a/packages/app/src/components/prompt-input/context-items.tsx b/packages/app/src/components/prompt-input/context-items.tsx index 6b154ee1e529..1f03f199ab8f 100644 --- a/packages/app/src/components/prompt-input/context-items.tsx +++ b/packages/app/src/components/prompt-input/context-items.tsx @@ -1,6 +1,8 @@ import { Component, For, Show } from "solid-js" +import { Dynamic } from "solid-js/web" import { FileIcon } from "@opencode-ai/ui/file-icon" import { IconButton } from "@opencode-ai/ui/icon-button" +import { Tooltip } from "@opencode-ai/ui/tooltip" import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" import { getDirectory, getFilename, getFilenameTruncated } from "@opencode-ai/core/util/path" import type { ContextItem } from "@/context/prompt" @@ -12,6 +14,7 @@ type ContextItemsProps = { active: (item: PromptContextItem) => boolean openComment: (item: PromptContextItem) => void remove: (item: PromptContextItem) => void + newLayoutDesigns: boolean t: (key: string) => string } @@ -27,10 +30,17 @@ export const PromptContextItems: Component = (props) => { const selected = props.active(item) return ( - - + {directory} {filename} @@ -78,7 +88,7 @@ export const PromptContextItems: Component = (props) => { {(comment) =>
{comment()}
}
- + ) }} diff --git a/packages/app/src/components/prompt-input/image-attachments.css b/packages/app/src/components/prompt-input/image-attachments.css new file mode 100644 index 000000000000..2aaf494ebfc4 --- /dev/null +++ b/packages/app/src/components/prompt-input/image-attachments.css @@ -0,0 +1,43 @@ +@keyframes prompt-attachments-fade-left { + from { + visibility: hidden; + } + to { + visibility: visible; + } +} + +@keyframes prompt-attachments-fade-right { + from { + visibility: visible; + } + to { + visibility: hidden; + } +} + +[data-slot="prompt-attachments"] { + timeline-scope: --prompt-attachments-scroll; + + [data-slot^="prompt-attachments-fade-"] { + visibility: hidden; + } +} + +@supports (animation-timeline: --prompt-attachments-scroll) and (timeline-scope: --prompt-attachments-scroll) { + [data-slot="prompt-attachments-scroll"] { + scroll-timeline: --prompt-attachments-scroll x; + } + + [data-slot="prompt-attachments-fade-left"] { + animation: prompt-attachments-fade-left linear both; + animation-timeline: --prompt-attachments-scroll; + animation-range: 0 0.1px; + } + + [data-slot="prompt-attachments-fade-right"] { + animation: prompt-attachments-fade-right linear both; + animation-timeline: --prompt-attachments-scroll; + animation-range: calc(100% - 1.1px) calc(100% - 1px); + } +} diff --git a/packages/app/src/components/prompt-input/image-attachments.tsx b/packages/app/src/components/prompt-input/image-attachments.tsx index dd8138e5a4f2..3a0cc3e4dd75 100644 --- a/packages/app/src/components/prompt-input/image-attachments.tsx +++ b/packages/app/src/components/prompt-input/image-attachments.tsx @@ -1,60 +1,167 @@ import { Component, For, Show } from "solid-js" import { Icon } from "@opencode-ai/ui/icon" +import { Icon as IconV2 } from "@opencode-ai/ui/v2/icon" import { Tooltip } from "@opencode-ai/ui/tooltip" -import type { ImageAttachmentPart } from "@/context/prompt" +import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" +import { AttachmentCardV2 } from "@opencode-ai/session-ui/v2/attachment-card-v2" +import { CommentCardV2 } from "@opencode-ai/session-ui/v2/comment-card-v2" +import { typeLabel } from "@opencode-ai/session-ui/message-file" +import type { ContextItem, ImageAttachmentPart } from "@/context/prompt" +import "./image-attachments.css" + +type PromptCommentItem = ContextItem & { key: string } type PromptImageAttachmentsProps = { attachments: ImageAttachmentPart[] onOpen: (attachment: ImageAttachmentPart) => void onRemove: (id: string) => void removeLabel: string + newLayoutDesigns: boolean + comments?: PromptCommentItem[] + commentActive?: (item: PromptCommentItem) => boolean + onOpenComment?: (item: PromptCommentItem) => void + onRemoveComment?: (item: PromptCommentItem) => void } const fallbackClass = "size-16 rounded-md bg-surface-base flex items-center justify-center border border-border-base" const imageClass = "size-16 rounded-md object-cover border border-border-base hover:border-border-strong-base transition-colors" +const imageClassV2 = "w-[58px] h-[46px] rounded-[6px] object-cover" +// inset box-shadows do not paint over content, so the hairline is a separate overlay +const imageHairlineClassV2 = + "absolute inset-0 rounded-[6px] shadow-[inset_0_0_0_0.5px_var(--v2-border-border-base)] pointer-events-none" const removeClass = "absolute -top-1.5 -right-1.5 size-5 rounded-full bg-surface-raised-stronger-non-alpha border border-border-base flex items-center justify-center opacity-0 group-hover:opacity-100 transition-opacity hover:bg-surface-raised-base-hover" +const removeClassV2 = + "absolute -top-1 -right-1 size-4 rounded-full bg-v2-icon-icon-muted outline-solid outline-1 outline-v2-icon-icon-contrast flex items-center justify-center opacity-0 group-hover:opacity-100 transition-opacity" const nameClass = "absolute bottom-0 left-0 right-0 px-1 py-0.5 bg-black/50 rounded-b-md" export const PromptImageAttachments: Component = (props) => { return ( - 0}> -
- - {(attachment) => ( - -
+ 0 || (props.newLayoutDesigns && (props.comments?.length ?? 0) > 0)}> +
+
+ + + {(item) => ( +
+ + props.onOpenComment?.(item)} + /> + + +
+ )} +
+
+ + {(attachment) => { + const image = attachment.mime.startsWith("image/") + const media = () => ( - -
+ + +
+ } + > + + {typeLabel(attachment.filename, attachment.mime)} + +
} > {attachment.filename} props.onOpen(attachment)} /> + ) + const name = () => ( +
+ {attachment.filename} +
+ ) + const remove = () => ( -
- {attachment.filename} -
-
-
- )} -
+ ) + // v2 keeps the remove button outside the tooltip trigger so hovering it dismisses the tooltip + return ( + +
+ {media()} + {name()} + {remove()} +
+ + } + > +
+ + {media()} + +
+ + + {remove()} +
+
+ ) + }} + +
+ +
+
+
) diff --git a/packages/app/src/components/prompt-input/submit.test.ts b/packages/app/src/components/prompt-input/submit.test.ts index 9a65fa101dd1..ad9fd31db811 100644 --- a/packages/app/src/components/prompt-input/submit.test.ts +++ b/packages/app/src/components/prompt-input/submit.test.ts @@ -1,5 +1,6 @@ import { beforeAll, beforeEach, describe, expect, mock, test } from "bun:test" import type { Prompt } from "@/context/prompt" +import type { ModelSelection } from "@/context/local" let createPromptSubmit: typeof import("./submit").createPromptSubmit @@ -33,6 +34,10 @@ const prompt = { current: () => promptValue, cursor: () => 0, dirty: () => true, + model: { + current: () => undefined, + set: () => undefined, + }, reset: () => undefined, set: () => undefined, context: { @@ -378,6 +383,39 @@ describe("prompt submit worktree selection", () => { }) }) + test("uses an injected model selection", async () => { + params = { id: "session-1" } + const model = { + current: () => ({ id: "draft-model", provider: { id: "draft-provider" } }), + variant: { current: () => "draft-variant" }, + } as unknown as ModelSelection + const submit = createPromptSubmit({ + prompt, + info: () => ({ id: "session-1" }), + imageAttachments: () => [], + commentCount: () => 0, + autoAccept: () => false, + mode: () => "normal", + working: () => false, + editor: () => undefined, + queueScroll: () => undefined, + promptLength: (value) => value.reduce((sum, part) => sum + ("content" in part ? part.content.length : 0), 0), + addToHistory: () => undefined, + resetHistoryNavigation: () => undefined, + setMode: () => undefined, + setPopover: () => undefined, + model, + }) + + await submit.handleSubmit({ preventDefault: () => undefined } as unknown as Event) + + expect(optimistic[0]).toMatchObject({ + message: { + model: { providerID: "draft-provider", modelID: "draft-model", variant: "draft-variant" }, + }, + }) + }) + test("seeds new sessions before optimistic prompts are added", async () => { const submit = createPromptSubmit({ prompt, diff --git a/packages/app/src/components/prompt-input/submit.ts b/packages/app/src/components/prompt-input/submit.ts index 05d978a947ce..c9f560d26929 100644 --- a/packages/app/src/components/prompt-input/submit.ts +++ b/packages/app/src/components/prompt-input/submit.ts @@ -3,12 +3,12 @@ import { showToast } from "@/utils/toast" import { base64Encode } from "@opencode-ai/core/util/encode" import { Binary } from "@opencode-ai/core/util/binary" import { useNavigate, useParams, useSearchParams } from "@solidjs/router" -import { batch, type Accessor } from "solid-js" +import { batch, startTransition, type Accessor } from "solid-js" import { useTabs } from "@/context/tabs" import { useServerSync, type ServerSync } from "@/context/server-sync" import { useLanguage } from "@/context/language" import { useLayout } from "@/context/layout" -import { useLocal } from "@/context/local" +import { useLocal, type ModelSelection } from "@/context/local" import { usePermission } from "@/context/permission" import { type ContextItem, type ImageAttachmentPart, type Prompt, type usePrompt } from "@/context/prompt" import { useSDK, type DirectorySDK } from "@/context/sdk" @@ -191,6 +191,7 @@ type PromptSubmitInput = { onQueue?: (draft: FollowupDraft) => void onAbort?: () => void onSubmit?: () => void + model?: ModelSelection } export function createPromptSubmit(input: PromptSubmitInput) { @@ -298,9 +299,10 @@ export function createPromptSubmit(input: PromptSubmitInput) { return } - const currentModel = local.model.current() + const modelSelection = input.model ?? local.model + const currentModel = modelSelection.current() const currentAgent = local.agent.current() - const variant = local.model.variant.current() + const variant = modelSelection.variant.current() if (!currentModel || !currentAgent) { showToast({ title: language.t("prompt.toast.modelAgentRequired.title"), @@ -374,13 +376,20 @@ export function createPromptSubmit(input: PromptSubmitInput) { if (created) { seed(sessionDirectory, created) session = created - if (shouldAutoAccept) permission.enableAutoAccept(session.id, sessionDirectory) - local.session.promote(sessionDirectory, session.id) - layout.handoff.setTabs(base64Encode(sessionDirectory), session.id) - const draftID = search.draftId - if (draftID) tabs.promoteDraft(draftID, { server: tabs.draft(draftID).server, sessionId: session.id }) - else navigate(`/${base64Encode(sessionDirectory)}/session/${session.id}`) - submission.retarget(prompt.capture({ dir: base64Encode(sessionDirectory), id: session.id })) + await startTransition(() => { + if (!session) return + if (shouldAutoAccept) permission.enableAutoAccept(session.id, sessionDirectory) + local.session.promote(sessionDirectory, session.id, { + agent: currentAgent.name, + model: { providerID: currentModel.provider.id, modelID: currentModel.id }, + variant: variant ?? null, + }) + layout.handoff.setTabs(base64Encode(sessionDirectory), session.id) + const draftID = search.draftId + if (draftID) tabs.promoteDraft(draftID, { server: tabs.draft(draftID).server, sessionId: session.id }) + else navigate(`/${base64Encode(sessionDirectory)}/session/${session.id}`) + submission.retarget(prompt.capture({ dir: base64Encode(sessionDirectory), id: session.id })) + }) } } if (!session) { diff --git a/packages/app/src/components/prompt-workspace-selector.tsx b/packages/app/src/components/prompt-workspace-selector.tsx index 27d8c97ca2a7..35ac4f9c6edd 100644 --- a/packages/app/src/components/prompt-workspace-selector.tsx +++ b/packages/app/src/components/prompt-workspace-selector.tsx @@ -1,5 +1,6 @@ import { For, Show } from "solid-js" import { MenuV2 } from "@opencode-ai/ui/v2/menu-v2" +import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" import { Icon } from "@opencode-ai/ui/icon" import { Icon as IconV2 } from "@opencode-ai/ui/v2/icon" import { getFilename } from "@opencode-ai/core/util/path" @@ -96,10 +97,17 @@ export function PromptWorkspaceSelector(props: { {(branch) => ( <> -
- - {branch()} -
+ +
+ + {branch()} +
+
)} diff --git a/packages/app/src/components/session-context-usage.tsx b/packages/app/src/components/session-context-usage.tsx index fd2d1dd1a30e..c1e0ccde0597 100644 --- a/packages/app/src/components/session-context-usage.tsx +++ b/packages/app/src/components/session-context-usage.tsx @@ -12,7 +12,7 @@ import { useSync } from "@/context/sync" import { useLanguage } from "@/context/language" import { useProviders } from "@/hooks/use-providers" import { useSDK } from "@/context/sdk" -import { getSessionContext, getSessionTokenTotal } from "@/components/session/session-context-metrics" +import { getSessionContext } from "@/components/session/session-context-metrics" import { useSessionLayout } from "@/pages/session/session-layout" import { createSessionTabs } from "@/pages/session/helpers" import { useSettings } from "@/context/settings" @@ -74,7 +74,6 @@ export function SessionContextUsage(props: SessionContextUsageProps) { ) const context = createMemo(() => getSessionContext(messages(), [...providers.all().values()])) - const tokens = createMemo(() => info()?.tokens) const cost = createMemo(() => { return usd().format(info()?.cost ?? 0) }) @@ -132,7 +131,7 @@ export function SessionContextUsage(props: SessionContextUsageProps) {
) diff --git a/packages/app/src/components/session/index.ts b/packages/app/src/components/session/index.ts index 8e424f0f36f7..9e44cea17be6 100644 --- a/packages/app/src/components/session/index.ts +++ b/packages/app/src/components/session/index.ts @@ -1,6 +1,7 @@ export { SessionHeader } from "./session-header" export { SessionContextTab } from "./session-context-tab" export { SortableTab, FileVisual } from "./session-sortable-tab" +export { SortableTabV2 } from "./session-sortable-tab-v2" export { SortableTerminalTab } from "./session-sortable-terminal-tab" export { NewSessionView } from "./session-new-view" export { NewSessionDesignView } from "./session-new-design-view" diff --git a/packages/app/src/components/session/open-in-app-v2.tsx b/packages/app/src/components/session/open-in-app-v2.tsx new file mode 100644 index 000000000000..e26ff2e0eaba --- /dev/null +++ b/packages/app/src/components/session/open-in-app-v2.tsx @@ -0,0 +1,98 @@ +import { For, Show } from "solid-js" +import { AppIcon } from "@opencode-ai/ui/app-icon" +import { Icon } from "@opencode-ai/ui/icon" +import { Spinner } from "@opencode-ai/ui/spinner" +import { Icon as IconV2 } from "@opencode-ai/ui/v2/icon" +import { MenuV2 } from "@opencode-ai/ui/v2/menu-v2" +import { SplitButtonV2, SplitButtonV2Action, SplitButtonV2MenuTrigger } from "@opencode-ai/ui/v2/split-button-v2" +import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" +import { useLanguage } from "@/context/language" +import { type OpenApp, useOpenInApp } from "@/components/session/open-in-app" + +export function OpenInAppV2(props: { directory: () => string }) { + const language = useLanguage() + const state = useOpenInApp(props) + + return ( + + event.stopPropagation()}> + + event.stopPropagation()} + onClick={(event) => { + event.stopPropagation() + if (state.opening()) return + state.openDir(state.current().id) + }} + disabled={state.opening()} + aria-label={language.t("session.header.open.ariaLabel", { app: state.current().label })} + > + }> + + + + + state.setMenu("open", open)} + > + event.stopPropagation()} + > + + + + + + {language.t("session.header.openIn")} + { + state.selectApp(value as OpenApp) + }} + > + + {(option) => ( + { + state.selectApp(option.id) + state.setMenu("open", false) + state.openDir(option.id) + }} + > + + {option.label} + + )} + + + + + { + state.setMenu("open", false) + state.copyPath() + }} + > + + {language.t("session.header.open.copyPath")} + + + + + + + ) +} diff --git a/packages/app/src/components/session/open-in-app.tsx b/packages/app/src/components/session/open-in-app.tsx new file mode 100644 index 000000000000..cd363d00c49a --- /dev/null +++ b/packages/app/src/components/session/open-in-app.tsx @@ -0,0 +1,229 @@ +import { createEffect, createMemo } from "solid-js" +import { createStore } from "solid-js/store" +import { useLanguage } from "@/context/language" +import { usePlatform } from "@/context/platform" +import { useServer } from "@/context/server" +import { Persist, persisted } from "@/utils/persist" +import { showToast } from "@/utils/toast" + +export const OPEN_APPS = [ + "vscode", + "cursor", + "zed", + "textmate", + "antigravity", + "finder", + "terminal", + "iterm2", + "ghostty", + "warp", + "xcode", + "android-studio", + "powershell", + "sublime-text", +] as const + +export type OpenApp = (typeof OPEN_APPS)[number] +export type OpenAppOS = "macos" | "windows" | "linux" | "unknown" + +export const MAC_OPEN_APPS = [ + { + id: "vscode", + label: "session.header.open.app.vscode", + icon: "vscode", + openWith: "Visual Studio Code", + }, + { id: "cursor", label: "session.header.open.app.cursor", icon: "cursor", openWith: "Cursor" }, + { id: "zed", label: "session.header.open.app.zed", icon: "zed", openWith: "Zed" }, + { id: "textmate", label: "session.header.open.app.textmate", icon: "textmate", openWith: "TextMate" }, + { + id: "antigravity", + label: "session.header.open.app.antigravity", + icon: "antigravity", + openWith: "Antigravity", + }, + { id: "terminal", label: "session.header.open.app.terminal", icon: "terminal", openWith: "Terminal" }, + { id: "iterm2", label: "session.header.open.app.iterm2", icon: "iterm2", openWith: "iTerm" }, + { id: "ghostty", label: "session.header.open.app.ghostty", icon: "ghostty", openWith: "Ghostty" }, + { id: "warp", label: "session.header.open.app.warp", icon: "warp", openWith: "Warp" }, + { id: "xcode", label: "session.header.open.app.xcode", icon: "xcode", openWith: "Xcode" }, + { + id: "android-studio", + label: "session.header.open.app.androidStudio", + icon: "android-studio", + openWith: "Android Studio", + }, + { + id: "sublime-text", + label: "session.header.open.app.sublimeText", + icon: "sublime-text", + openWith: "Sublime Text", + }, +] as const + +export const WINDOWS_OPEN_APPS = [ + { id: "vscode", label: "session.header.open.app.vscode", icon: "vscode", openWith: "code" }, + { id: "cursor", label: "session.header.open.app.cursor", icon: "cursor", openWith: "cursor" }, + { id: "zed", label: "session.header.open.app.zed", icon: "zed", openWith: "zed" }, + { + id: "powershell", + label: "session.header.open.app.powershell", + icon: "powershell", + openWith: "powershell", + }, + { + id: "sublime-text", + label: "session.header.open.app.sublimeText", + icon: "sublime-text", + openWith: "Sublime Text", + }, +] as const + +export const LINUX_OPEN_APPS = [ + { id: "vscode", label: "session.header.open.app.vscode", icon: "vscode", openWith: "code" }, + { id: "cursor", label: "session.header.open.app.cursor", icon: "cursor", openWith: "cursor" }, + { id: "zed", label: "session.header.open.app.zed", icon: "zed", openWith: "zed" }, + { + id: "sublime-text", + label: "session.header.open.app.sublimeText", + icon: "sublime-text", + openWith: "Sublime Text", + }, +] as const + +export function detectOpenAppOS(platform: ReturnType): OpenAppOS { + if (platform.platform === "desktop" && platform.os) return platform.os + if (typeof navigator !== "object") return "unknown" + const value = navigator.platform || navigator.userAgent + if (/Mac/i.test(value)) return "macos" + if (/Win/i.test(value)) return "windows" + if (/Linux/i.test(value)) return "linux" + return "unknown" +} + +export function openAppFileManager(os: OpenAppOS) { + if (os === "macos") return { label: "session.header.open.finder", icon: "finder" as const } + if (os === "windows") return { label: "session.header.open.fileExplorer", icon: "file-explorer" as const } + return { label: "session.header.open.fileManager", icon: "finder" as const } +} + +export function openAppsForOS(os: OpenAppOS) { + if (os === "macos") return MAC_OPEN_APPS + if (os === "windows") return WINDOWS_OPEN_APPS + return LINUX_OPEN_APPS +} + +const showRequestError = (language: ReturnType, err: unknown) => { + showToast({ + variant: "error", + title: language.t("common.requestFailed"), + description: err instanceof Error ? err.message : String(err), + }) +} + +export function useOpenInApp(input: { directory: () => string }) { + const platform = usePlatform() + const server = useServer() + const language = useLanguage() + + const os = createMemo(() => detectOpenAppOS(platform)) + const apps = createMemo(() => openAppsForOS(os())) + const fileManager = createMemo(() => openAppFileManager(os())) + + const [exists, setExists] = createStore>>({ + finder: true, + }) + + createEffect(() => { + if (platform.platform !== "desktop") return + if (!platform.checkAppExists) return + + const list = apps() + + setExists(Object.fromEntries(list.map((app) => [app.id, undefined])) as Partial>) + + void Promise.all( + list.map((app) => + Promise.resolve(platform.checkAppExists?.(app.openWith)) + .then((value) => Boolean(value)) + .catch(() => false) + .then((ok) => [app.id, ok] as const), + ), + ).then((entries) => { + setExists(Object.fromEntries(entries) as Partial>) + }) + }) + + const options = createMemo(() => { + return [ + { id: "finder", label: language.t(fileManager().label), icon: fileManager().icon }, + ...apps() + .filter((app) => exists[app.id]) + .map((app) => ({ ...app, label: language.t(app.label) })), + ] as const + }) + + const [prefs, setPrefs] = persisted(Persist.global("open.app"), createStore({ app: "finder" as OpenApp | "finder" })) + const [menu, setMenu] = createStore({ open: false }) + const [openRequest, setOpenRequest] = createStore({ + app: undefined as OpenApp | undefined, + }) + + const canOpen = createMemo(() => platform.platform === "desktop" && !!platform.openPath && server.isLocal()) + const current = createMemo( + () => + options().find((o) => o.id === prefs.app) ?? + options()[0] ?? + ({ id: "finder", label: fileManager().label, icon: fileManager().icon } as const), + ) + const opening = createMemo(() => openRequest.app !== undefined) + + const selectApp = (app: OpenApp | "finder") => { + if (!options().some((item) => item.id === app)) return + setPrefs("app", app) + } + + const openDir = (app: OpenApp | "finder") => { + if (opening() || !canOpen() || !platform.openPath) return + const directory = input.directory() + if (!directory) return + + const item = options().find((o) => o.id === app) + const openWith = item && "openWith" in item ? item.openWith : undefined + setOpenRequest("app", app) + platform + .openPath(directory, openWith) + .catch((err: unknown) => showRequestError(language, err)) + .finally(() => { + setOpenRequest("app", undefined) + }) + } + + const copyPath = () => { + const directory = input.directory() + if (!directory) return + navigator.clipboard + .writeText(directory) + .then(() => { + showToast({ + variant: "success", + icon: "circle-check", + title: language.t("session.share.copy.copied"), + description: directory, + }) + }) + .catch((err: unknown) => showRequestError(language, err)) + } + + return { + canOpen, + opening, + current, + options, + menu, + setMenu, + openDir, + selectApp, + copyPath, + } +} diff --git a/packages/app/src/components/session/session-context-metrics.test.ts b/packages/app/src/components/session/session-context-metrics.test.ts index 5b11bf143393..37cc0d5b5d2e 100644 --- a/packages/app/src/components/session/session-context-metrics.test.ts +++ b/packages/app/src/components/session/session-context-metrics.test.ts @@ -1,6 +1,6 @@ import { describe, expect, test } from "bun:test" import type { Message } from "@opencode-ai/sdk/v2/client" -import { getSessionContext, getSessionTokenTotal } from "./session-context-metrics" +import { getSessionContext } from "./session-context-metrics" const assistant = ( id: string, @@ -38,10 +38,10 @@ const user = (id: string) => { } describe("getSessionContext", () => { - test("computes usage from latest assistant with tokens", () => { + test("computes token totals and usage from latest assistant with tokens", () => { const messages = [ user("u1"), - assistant("a1", { input: 0, output: 0, reasoning: 0, read: 0, write: 0 }, 0.5), + assistant("a1", { input: 600, output: 200, reasoning: 100, read: 50, write: 50 }, 0.5), assistant("a2", { input: 300, output: 100, reasoning: 50, read: 25, write: 25 }, 1.25), ] const providers = [ @@ -60,6 +60,8 @@ describe("getSessionContext", () => { const ctx = getSessionContext(messages, providers) expect(ctx?.message.id).toBe("a2") + expect(ctx?.total).toBe(500) + expect(ctx?.input).toBe(300) expect(ctx?.usage).toBe(50) expect(ctx?.providerLabel).toBe("OpenAI") expect(ctx?.modelLabel).toBe("GPT-4.1") @@ -94,15 +96,4 @@ describe("getSessionContext", () => { expect(ctx).toBeUndefined() }) - - test("computes stored session token totals", () => { - expect( - getSessionTokenTotal({ - input: 10, - output: 20, - reasoning: 30, - cache: { read: 40, write: 50 }, - }), - ).toBe(150) - }) }) diff --git a/packages/app/src/components/session/session-context-metrics.ts b/packages/app/src/components/session/session-context-metrics.ts index 84cc984abfab..30dc9e958b9c 100644 --- a/packages/app/src/components/session/session-context-metrics.ts +++ b/packages/app/src/components/session/session-context-metrics.ts @@ -1,4 +1,4 @@ -import type { AssistantMessage, Message, Session } from "@opencode-ai/sdk/v2/client" +import type { AssistantMessage, Message } from "@opencode-ai/sdk/v2/client" type Provider = { id: string @@ -21,6 +21,7 @@ type Context = { modelLabel: string limit: number | undefined input: number + total: number usage: number | null } @@ -54,6 +55,7 @@ const build = (messages: Message[] = [], providers: Provider[] = []): Context | modelLabel: model?.name ?? message.modelID, limit, input: message.tokens.input, + total, usage: limit ? Math.round((total / limit) * 100) : null, } } @@ -61,8 +63,3 @@ const build = (messages: Message[] = [], providers: Provider[] = []): Context | export function getSessionContext(messages: Message[] = [], providers: Provider[] = []) { return build(messages, providers) } - -export function getSessionTokenTotal(tokens: Session["tokens"] | undefined) { - if (!tokens) return undefined - return tokens.input + tokens.output + tokens.reasoning + tokens.cache.read + tokens.cache.write -} diff --git a/packages/app/src/components/session/session-context-tab.tsx b/packages/app/src/components/session/session-context-tab.tsx index 8757c693678b..25ad5ab230ca 100644 --- a/packages/app/src/components/session/session-context-tab.tsx +++ b/packages/app/src/components/session/session-context-tab.tsx @@ -15,7 +15,7 @@ import { useLanguage } from "@/context/language" import { useProviders } from "@/hooks/use-providers" import { useSDK } from "@/context/sdk" import { useSessionLayout } from "@/pages/session/session-layout" -import { getSessionContext, getSessionTokenTotal } from "./session-context-metrics" +import { getSessionContext } from "./session-context-metrics" import { estimateSessionContextBreakdown, type SessionContextBreakdownKey } from "./session-context-breakdown" import { createSessionContextFormatter } from "./session-context-format" @@ -135,7 +135,6 @@ export function SessionContextTab() { ) const ctx = createMemo(() => getSessionContext(messages(), [...providers.all().values()])) - const tokens = createMemo(() => info()?.tokens) const formatter = createMemo(() => createSessionContextFormatter(language.intl())) const cost = createMemo(() => { @@ -204,14 +203,15 @@ export function SessionContextTab() { { label: "context.stats.provider", value: providerLabel }, { label: "context.stats.model", value: modelLabel }, { label: "context.stats.limit", value: () => formatter().number(ctx()?.limit) }, - { label: "context.stats.totalTokens", value: () => formatter().number(getSessionTokenTotal(tokens())) }, + { label: "context.stats.totalTokens", value: () => formatter().number(ctx()?.total) }, { label: "context.stats.usage", value: () => formatter().percent(ctx()?.usage) }, - { label: "context.stats.inputTokens", value: () => formatter().number(tokens()?.input) }, - { label: "context.stats.outputTokens", value: () => formatter().number(tokens()?.output) }, - { label: "context.stats.reasoningTokens", value: () => formatter().number(tokens()?.reasoning) }, + { label: "context.stats.inputTokens", value: () => formatter().number(ctx()?.input) }, + { label: "context.stats.outputTokens", value: () => formatter().number(ctx()?.message.tokens.output) }, + { label: "context.stats.reasoningTokens", value: () => formatter().number(ctx()?.message.tokens.reasoning) }, { label: "context.stats.cacheTokens", - value: () => `${formatter().number(tokens()?.cache.read)} / ${formatter().number(tokens()?.cache.write)}`, + value: () => + `${formatter().number(ctx()?.message.tokens.cache.read)} / ${formatter().number(ctx()?.message.tokens.cache.write)}`, }, { label: "context.stats.userMessages", value: () => counts().user.toLocaleString(language.intl()) }, { label: "context.stats.assistantMessages", value: () => counts().assistant.toLocaleString(language.intl()) }, diff --git a/packages/app/src/components/session/session-sortable-tab-v2.tsx b/packages/app/src/components/session/session-sortable-tab-v2.tsx new file mode 100644 index 000000000000..158eba9f0b68 --- /dev/null +++ b/packages/app/src/components/session/session-sortable-tab-v2.tsx @@ -0,0 +1,66 @@ +import { createMemo, Show } from "solid-js" +import type { JSX } from "solid-js" +import { useSortable } from "@dnd-kit/solid/sortable" +import { IconButton } from "@opencode-ai/ui/icon-button" +import { TooltipKeybind } from "@opencode-ai/ui/tooltip" +import { Tabs } from "@opencode-ai/ui/tabs" +import { useFile } from "@/context/file" +import { useLanguage } from "@/context/language" +import { useCommand } from "@/context/command" +import { FileVisual } from "./session-sortable-tab" + +export function SortableTabV2(props: { + tab: string + index: () => number + temporary?: boolean + onTabClose: (tab: string) => void + onTabDoubleClick?: (tab: string) => void +}): JSX.Element { + const file = useFile() + const language = useLanguage() + const command = useCommand() + const sortable = useSortable({ + get id() { + return props.tab + }, + get index() { + return props.index() + }, + }) + const path = createMemo(() => file.pathFromTab(props.tab)) + const content = createMemo(() => { + const value = path() + if (!value) return + return + }) + return ( +
+
+ + props.onTabClose(props.tab)} + aria-label={language.t("common.closeTab")} + /> + + } + hideCloseButton + onMiddleClick={() => props.onTabClose(props.tab)} + onDblClick={() => props.onTabDoubleClick?.(props.tab)} + > + {(value) => value()} + +
+
+ ) +} diff --git a/packages/app/src/components/session/session-sortable-terminal-tab-v2.tsx b/packages/app/src/components/session/session-sortable-terminal-tab-v2.tsx index f0cdf41ac1f5..301783a0dfa8 100644 --- a/packages/app/src/components/session/session-sortable-terminal-tab-v2.tsx +++ b/packages/app/src/components/session/session-sortable-terminal-tab-v2.tsx @@ -211,7 +211,17 @@ export function SortableTerminalTabV2(props: { { + // Switch on mousedown to shave the press-release delay off tab switches. + if (e.button !== 0) return + if (store.editing) return + focus() + }} + onClick={(e) => { + // Mouse navigation already happened on mousedown; detail 0 means keyboard activation. + if (e.detail > 0) return + focus() + }} closeButton={ span { diff --git a/packages/app/src/components/terminal.tsx b/packages/app/src/components/terminal.tsx index 8998fc06c89e..c5faf85bff3e 100644 --- a/packages/app/src/components/terminal.tsx +++ b/packages/app/src/components/terminal.tsx @@ -1,7 +1,8 @@ import { withAlpha } from "@opencode-ai/ui/theme/color" import { useTheme } from "@opencode-ai/ui/theme/context" import { resolveThemeVariant } from "@opencode-ai/ui/theme/resolve" -import type { HexColor } from "@opencode-ai/ui/theme/types" +import { resolveThemeVariantV2 } from "@opencode-ai/ui/theme/v2/resolve" +import type { HexColor, ResolvedV2Theme } from "@opencode-ai/ui/theme/types" import { showToast } from "@/utils/toast" import type { FitAddon, Ghostty, Terminal as Term } from "ghostty-web" import { type ComponentProps, createEffect, createMemo, onCleanup, onMount, splitProps } from "solid-js" @@ -68,6 +69,19 @@ const debugTerminal = (...values: unknown[]) => { console.debug("[terminal]", ...values) } +const resolveV2Token = (tokens: ResolvedV2Theme, key: string) => { + let current = tokens[key] + for (let i = 0; i < 8 && current; i++) { + const match = /^var\(--([^)]+)\)$/.exec(current.trim()) + if (!match) { + const hex = current.trim() + if (/^#[0-9a-fA-F]{8}$/.test(hex)) return hex.slice(0, 7) + return hex + } + current = tokens[match[1]] + } +} + const useTerminalUiBindings = (input: { container: HTMLDivElement term: Term @@ -238,7 +252,10 @@ export const Terminal = (props: TerminalProps) => { if (!variant?.seeds && !variant?.palette) return fallback const resolved = resolveThemeVariant(variant, mode === "dark") const text = resolved["text-stronger"] ?? fallback.foreground - const background = resolved["background-stronger"] ?? fallback.background + const background = settings.general.newLayoutDesigns() + ? (resolveV2Token(resolveThemeVariantV2(variant, mode === "dark"), "v2-background-bg-base") ?? + fallback.background) + : (resolved["background-stronger"] ?? fallback.background) const alpha = mode === "dark" ? 0.25 : 0.2 const base = text.startsWith("#") ? (text as HexColor) : (fallback.foreground as HexColor) const selectionBackground = withAlpha(base, alpha) diff --git a/packages/app/src/components/titlebar.tsx b/packages/app/src/components/titlebar.tsx index 2a818d565da1..994d69a60fb2 100644 --- a/packages/app/src/components/titlebar.tsx +++ b/packages/app/src/components/titlebar.tsx @@ -25,6 +25,7 @@ import { readSessionTabsRemovedDetail, SESSION_TABS_REMOVED_EVENT } from "@/comp import { useGlobal } from "@/context/global" import { ServerConnection, useServer } from "@/context/server" import { tabKey, useTabs } from "@/context/tabs" +import type { PromptSession } from "@/context/prompt" import "./titlebar.css" import { newTabTooltipKeybind } from "./command-tooltip-keybind" @@ -324,13 +325,20 @@ export function Titlebar(props: { update?: TitlebarUpdate }) { const route = layout.route() const activeSession = session() if (route.type === "session" && activeSession) { - tabs.newDraft({ server: route.server ?? server.key, directory: activeSession.directory }, "") + const sessionTab = { + type: "session" as const, + server: route.server ?? server.key, + sessionId: activeSession.id, + } + const model = tabs.stateValue(sessionTab, "prompt")?.model.current() + tabs.newDraft({ server: sessionTab.server, directory: activeSession.directory }, "", model) return } const activeTab = currentTab() if (activeTab?.type === "draft") { - tabs.newDraft({ server: activeTab.server, directory: activeTab.directory }, "") + const model = tabs.stateValue(activeTab, "prompt")?.model.current() + tabs.newDraft({ server: activeTab.server, directory: activeTab.directory }, "", model) return } diff --git a/packages/app/src/context/layout.tsx b/packages/app/src/context/layout.tsx index dcac3e778104..7ac248dc1d1b 100644 --- a/packages/app/src/context/layout.tsx +++ b/packages/app/src/context/layout.tsx @@ -47,12 +47,20 @@ export function getAvatarColors(key?: string) { } export function getProjectAvatarVariant(key?: string): ProjectAvatarVariant { - if (key === "orange") return "orange" - if (key === "pink") return "pink" - if (key === "cyan") return "cyan" - if (key === "purple") return "purple" if (key === "mint") return "cyan" if (key === "lime") return "green" + if ( + key === "orange" || + key === "yellow" || + key === "cyan" || + key === "green" || + key === "red" || + key === "pink" || + key === "blue" || + key === "purple" || + key === "gray" + ) + return key return "gray" } diff --git a/packages/app/src/context/local.tsx b/packages/app/src/context/local.tsx index 415da747e0be..a12fd832fd1c 100644 --- a/packages/app/src/context/local.tsx +++ b/packages/app/src/context/local.tsx @@ -67,7 +67,7 @@ export const { use: useLocal, provider: LocalProvider } = createSimpleContext({ const list = createMemo(() => sync().data.agent.filter((item) => item.mode !== "subagent" && !item.hidden)) const connected = createMemo(() => new Set(providers.connected().map((item) => item.id))) - const [saved, setSaved] = persisted( + const [saved, setSaved, , savedReady] = persisted( { ...Persist.serverWorkspace(serverSDK().scope, sdk().directory, "model-selection", ["model-selection.v1"]), migrate, @@ -375,11 +375,12 @@ export const { use: useLocal, provider: LocalProvider } = createSimpleContext({ model, agent, session: { + ready: savedReady, reset() { setStore({ draft: undefined, promoting: undefined }) }, - promote(dir: string, session: string) { - const next = clone(snapshot()) + promote(dir: string, session: string, state?: State) { + const next = clone(state ?? snapshot()) if (!next) return const key = handoffKey(serverSDK().scope, dir, session) handoff.set(key, next) @@ -409,3 +410,5 @@ export const { use: useLocal, provider: LocalProvider } = createSimpleContext({ return result }, }) + +export type ModelSelection = ReturnType["model"] diff --git a/packages/app/src/context/platform.tsx b/packages/app/src/context/platform.tsx index e59bec9aafd9..1a061a667552 100644 --- a/packages/app/src/context/platform.tsx +++ b/packages/app/src/context/platform.tsx @@ -37,6 +37,9 @@ type PlatformBase = { /** Open a local path in a local app (desktop only) */ openPath?(path: string, app?: string): Promise + /** Reveal a local path in the system file manager; false when the path does not exist (desktop only) */ + revealPath?(path: string): Promise + /** Restart the app */ restart(): Promise diff --git a/packages/app/src/context/prompt-state.test.ts b/packages/app/src/context/prompt-state.test.ts new file mode 100644 index 000000000000..24afe5424e71 --- /dev/null +++ b/packages/app/src/context/prompt-state.test.ts @@ -0,0 +1,29 @@ +import { describe, expect, test } from "bun:test" +import { createRoot } from "solid-js" +import { createPromptState, DEFAULT_PROMPT } from "./prompt-state" + +describe("prompt state initialization", () => { + test("initializes prompt text, cursor, and model together", () => { + createRoot((dispose) => { + const model = { providerID: "anthropic", modelID: "claude", variant: "high" } + const prompt = createPromptState({ prompt: "hello", model }) + + expect(prompt.current()).toEqual([{ type: "text", content: "hello", start: 0, end: 5 }]) + expect(prompt.cursor()).toBe(5) + expect(prompt.model.current()).toEqual(model) + expect(prompt.model.current()).not.toBe(model) + dispose() + }) + }) + + test("uses the default prompt without initial values", () => { + createRoot((dispose) => { + const prompt = createPromptState() + + expect(prompt.current()).toEqual(DEFAULT_PROMPT) + expect(prompt.cursor()).toBeUndefined() + expect(prompt.model.current()).toBeUndefined() + dispose() + }) + }) +}) diff --git a/packages/app/src/context/prompt-state.ts b/packages/app/src/context/prompt-state.ts new file mode 100644 index 000000000000..462cfa638b8e --- /dev/null +++ b/packages/app/src/context/prompt-state.ts @@ -0,0 +1,265 @@ +import { checksum } from "@opencode-ai/core/util/encode" +import type { FilePartSource } from "@opencode-ai/sdk/v2/client" +import { batch, createMemo, type Accessor } from "solid-js" +import { createStore, type SetStoreFunction } from "solid-js/store" +import type { FileSelection } from "@/context/file" +import { Persist, persisted } from "@/utils/persist" +import type { ServerScope } from "@/utils/server-scope" + +interface PartBase { + content: string + start: number + end: number +} + +export interface TextPart extends PartBase { + type: "text" +} + +export interface FileAttachmentPart extends PartBase { + type: "file" + path: string + selection?: FileSelection + mime?: string + filename?: string + url?: string + source?: FilePartSource +} + +export interface AgentPart extends PartBase { + type: "agent" + name: string +} + +export interface ImageAttachmentPart { + type: "image" + id: string + filename: string + sourcePath?: string + mime: string + dataUrl: string +} + +export type ContentPart = TextPart | FileAttachmentPart | AgentPart | ImageAttachmentPart +export type Prompt = ContentPart[] + +export type PromptModel = { + providerID: string + modelID: string + variant?: string | null +} + +export type FileContextItem = { + type: "file" + path: string + selection?: FileSelection + comment?: string + commentID?: string + commentOrigin?: "review" | "file" + preview?: string +} + +export type ContextItem = FileContextItem +export type PromptScope = { draftID: string } | { dir: string; id?: string } + +export const DEFAULT_PROMPT: Prompt = [{ type: "text", content: "", start: 0, end: 0 }] + +type PromptStore = { + prompt: Prompt + cursor?: number + model?: PromptModel + context: { + items: (ContextItem & { key: string })[] + } +} + +type InitialPrompt = { + prompt?: string + model?: PromptModel +} + +function isSelectionEqual(a?: FileSelection, b?: FileSelection) { + if (!a && !b) return true + if (!a || !b) return false + return ( + a.startLine === b.startLine && a.startChar === b.startChar && a.endLine === b.endLine && a.endChar === b.endChar + ) +} + +function isPartEqual(partA: ContentPart, partB: ContentPart) { + switch (partA.type) { + case "text": + return partB.type === "text" && partA.content === partB.content + case "file": + return ( + partB.type === "file" && + partA.path === partB.path && + partA.mime === partB.mime && + partA.filename === partB.filename && + isSelectionEqual(partA.selection, partB.selection) + ) + case "agent": + return partB.type === "agent" && partA.name === partB.name + case "image": + return partB.type === "image" && partA.id === partB.id + } +} + +export function isPromptEqual(promptA: Prompt, promptB: Prompt): boolean { + if (promptA.length !== promptB.length) return false + for (let i = 0; i < promptA.length; i++) { + if (!isPartEqual(promptA[i], promptB[i])) return false + } + return true +} + +function cloneSelection(selection?: FileSelection) { + if (!selection) return undefined + return { ...selection } +} + +function clonePart(part: ContentPart): ContentPart { + if (part.type === "text") return { ...part } + if (part.type === "image") return { ...part } + if (part.type === "agent") return { ...part } + return { + ...part, + selection: cloneSelection(part.selection), + } +} + +function clonePrompt(prompt: Prompt): Prompt { + return prompt.map(clonePart) +} + +function contextItemKey(item: ContextItem) { + if (item.type !== "file") return item.type + const start = item.selection?.startLine + const end = item.selection?.endLine + const key = `${item.type}:${item.path}:${start}:${end}` + + if (item.commentID) return `${key}:c=${item.commentID}` + const comment = item.comment?.trim() + if (!comment) return key + const digest = checksum(comment) ?? comment + return `${key}:c=${digest.slice(0, 8)}` +} + +export function isCommentItem(item: ContextItem | (ContextItem & { key: string })) { + return item.type === "file" && !!item.comment?.trim() +} + +function createPromptActions(setStore: SetStoreFunction) { + return { + set(prompt: Prompt, cursorPosition?: number) { + const next = clonePrompt(prompt) + batch(() => { + setStore("prompt", next) + if (cursorPosition !== undefined) setStore("cursor", cursorPosition) + }) + }, + reset() { + batch(() => { + setStore("prompt", clonePrompt(DEFAULT_PROMPT)) + setStore("cursor", 0) + }) + }, + } +} + +function promptTarget(serverScope: ServerScope, scope: PromptScope) { + if ("draftID" in scope) return Persist.draft(scope.draftID, "prompt") + const legacy = `${scope.dir}/prompt${scope.id ? "/" + scope.id : ""}.v2` + return Persist.serverScoped(serverScope, scope.dir, scope.id, "prompt", [legacy]) +} + +function promptStore(initial?: InitialPrompt): PromptStore { + const text = initial?.prompt + return { + prompt: + text === undefined ? clonePrompt(DEFAULT_PROMPT) : [{ type: "text", content: text, start: 0, end: text.length }], + cursor: text === undefined ? undefined : text.length, + model: initial?.model ? { ...initial.model } : undefined, + context: { + items: [], + }, + } +} + +function createPromptStateValue(store: PromptStore, setStore: SetStoreFunction) { + const actions = createPromptActions(setStore) + const value = { + current: () => store.prompt, + cursor: createMemo(() => store.cursor), + dirty: () => !isPromptEqual(store.prompt, DEFAULT_PROMPT), + model: { + current: () => store.model, + set: (model: PromptModel | undefined) => setStore("model", model), + }, + context: { + items: createMemo(() => store.context.items), + add(item: ContextItem) { + const key = contextItemKey(item) + if (store.context.items.find((x) => x.key === key)) return + setStore("context", "items", (items) => [...items, { key, ...item }]) + }, + remove(key: string) { + setStore("context", "items", (items) => items.filter((x) => x.key !== key)) + }, + removeComment(path: string, commentID: string) { + setStore("context", "items", (items) => + items.filter((item) => !(item.type === "file" && item.path === path && item.commentID === commentID)), + ) + }, + updateComment(path: string, commentID: string, next: Partial & { comment?: string }) { + setStore("context", "items", (items) => + items.map((item) => { + if (item.type !== "file" || item.path !== path || item.commentID !== commentID) return item + const value = { ...item, ...next } + return { ...value, key: contextItemKey(value) } + }), + ) + }, + replaceComments(items: FileContextItem[]) { + setStore("context", "items", (current) => [ + ...current.filter((item) => !isCommentItem(item)), + ...items.map((item) => ({ ...item, key: contextItemKey(item) })), + ]) + }, + }, + set: actions.set, + reset: actions.reset, + capture: () => value, + } + return value +} + +function createPersistedPrompt(target: ReturnType, initial?: InitialPrompt) { + const [store, setStore, _, ready] = persisted(target, createStore(promptStore(initial))) + return { ready, ...createPromptStateValue(store, setStore) } +} + +export function createPromptSession(serverScope: ServerScope, scope: PromptScope, initial?: InitialPrompt) { + return createPersistedPrompt(promptTarget(serverScope, scope), initial) +} + +export function createDraftPromptSession(draftID: string, initial?: InitialPrompt) { + return createPersistedPrompt(Persist.draft(draftID, "prompt"), initial) +} + +export type PromptSession = ReturnType + +export function createPromptReady(session: Accessor) { + return Object.defineProperty(() => session().ready(), "promise", { + get: () => session().ready.promise, + }) as (() => boolean) & { readonly promise: Promise | undefined } +} + +export function createPromptState(initial?: InitialPrompt) { + const [store, setStore] = createStore(promptStore(initial)) + const ready = Object.assign(() => true, { promise: Promise.resolve(true) }) + return { + ready, + ...createPromptStateValue(store, setStore), + } +} diff --git a/packages/app/src/context/prompt.tsx b/packages/app/src/context/prompt.tsx index c36149351624..14147b6616f0 100644 --- a/packages/app/src/context/prompt.tsx +++ b/packages/app/src/context/prompt.tsx @@ -1,186 +1,50 @@ +import { base64Encode } from "@opencode-ai/core/util/encode" import { createSimpleContext } from "@opencode-ai/ui/context" -import { base64Encode, checksum } from "@opencode-ai/core/util/encode" import { useParams, useSearchParams } from "@solidjs/router" -import { batch, createMemo, createRoot, getOwner, onCleanup, type Accessor } from "solid-js" -import { createStore, type SetStoreFunction } from "solid-js/store" -import type { FileSelection } from "@/context/file" -import { Persist, persisted } from "@/utils/persist" +import { createMemo, createRoot, getOwner, onCleanup } from "solid-js" +import { requireServerKey } from "@/utils/session-route" +import { ServerConnection } from "./server" import { useServerSDK } from "./server-sdk" -import type { ServerScope } from "@/utils/server-scope" +import { useSettings } from "./settings" import { useSDK } from "./sdk" import { useTabs, type Tab } from "./tabs" -import { ServerConnection } from "./server" -import { requireServerKey } from "@/utils/session-route" -import { useSettings } from "./settings" -import type { FilePartSource } from "@opencode-ai/sdk/v2/client" - -interface PartBase { - content: string - start: number - end: number -} - -export interface TextPart extends PartBase { - type: "text" -} - -export interface FileAttachmentPart extends PartBase { - type: "file" - path: string - selection?: FileSelection - mime?: string - filename?: string - url?: string - source?: FilePartSource -} - -export interface AgentPart extends PartBase { - type: "agent" - name: string -} - -export interface ImageAttachmentPart { - type: "image" - id: string - filename: string - sourcePath?: string - mime: string - dataUrl: string -} - -export type ContentPart = TextPart | FileAttachmentPart | AgentPart | ImageAttachmentPart -export type Prompt = ContentPart[] - -export type FileContextItem = { - type: "file" - path: string - selection?: FileSelection - comment?: string - commentID?: string - commentOrigin?: "review" | "file" - preview?: string -} - -export type ContextItem = FileContextItem - -export const DEFAULT_PROMPT: Prompt = [{ type: "text", content: "", start: 0, end: 0 }] - -function isSelectionEqual(a?: FileSelection, b?: FileSelection) { - if (!a && !b) return true - if (!a || !b) return false - return ( - a.startLine === b.startLine && a.startChar === b.startChar && a.endLine === b.endLine && a.endChar === b.endChar - ) -} - -function isPartEqual(partA: ContentPart, partB: ContentPart) { - switch (partA.type) { - case "text": - return partB.type === "text" && partA.content === partB.content - case "file": - return ( - partB.type === "file" && - partA.path === partB.path && - partA.mime === partB.mime && - partA.filename === partB.filename && - isSelectionEqual(partA.selection, partB.selection) - ) - case "agent": - return partB.type === "agent" && partA.name === partB.name - case "image": - return partB.type === "image" && partA.id === partB.id - } -} - -export function isPromptEqual(promptA: Prompt, promptB: Prompt): boolean { - if (promptA.length !== promptB.length) return false - for (let i = 0; i < promptA.length; i++) { - if (!isPartEqual(promptA[i], promptB[i])) return false - } - return true -} - -function cloneSelection(selection?: FileSelection) { - if (!selection) return undefined - return { ...selection } -} - -function clonePart(part: ContentPart): ContentPart { - if (part.type === "text") return { ...part } - if (part.type === "image") return { ...part } - if (part.type === "agent") return { ...part } - return { - ...part, - selection: cloneSelection(part.selection), - } -} - -function clonePrompt(prompt: Prompt): Prompt { - return prompt.map(clonePart) -} - -function contextItemKey(item: ContextItem) { - if (item.type !== "file") return item.type - const start = item.selection?.startLine - const end = item.selection?.endLine - const key = `${item.type}:${item.path}:${start}:${end}` - - if (item.commentID) { - return `${key}:c=${item.commentID}` - } - - const comment = item.comment?.trim() - if (!comment) return key - const digest = checksum(comment) ?? comment - return `${key}:c=${digest.slice(0, 8)}` -} - -function isCommentItem(item: ContextItem | (ContextItem & { key: string })) { - return item.type === "file" && !!item.comment?.trim() -} - -function createPromptActions( - setStore: SetStoreFunction<{ - prompt: Prompt - cursor?: number - context: { - items: (ContextItem & { key: string })[] - } - }>, -) { - return { - set(prompt: Prompt, cursorPosition?: number) { - const next = clonePrompt(prompt) - batch(() => { - setStore("prompt", next) - if (cursorPosition !== undefined) setStore("cursor", cursorPosition) - }) - }, - reset() { - batch(() => { - setStore("prompt", clonePrompt(DEFAULT_PROMPT)) - setStore("cursor", 0) - }) - }, - } -} +import { + createPromptReady, + createPromptSession, + type ContextItem, + type FileContextItem, + type Prompt, + type PromptModel, + type PromptScope, + type PromptSession, +} from "./prompt-state" + +export { + createPromptReady, + createPromptSession, + createPromptState, + DEFAULT_PROMPT, + isCommentItem, + isPromptEqual, +} from "./prompt-state" +export type { + AgentPart, + ContentPart, + ContextItem, + FileAttachmentPart, + FileContextItem, + ImageAttachmentPart, + Prompt, + PromptModel, + PromptScope, + PromptSession, + TextPart, +} from "./prompt-state" const WORKSPACE_KEY = "__workspace__" const MAX_PROMPT_SESSIONS = 20 -type PromptSession = ReturnType - -type PromptStore = { - prompt: Prompt - cursor?: number - context: { - items: (ContextItem & { key: string })[] - } -} - -type Scope = { draftID: string } | { dir: string; id?: string } - -export function selectPromptTab(tabs: Tab[], scope: Scope, server: ServerConnection.Key) { +export function selectPromptTab(tabs: Tab[], scope: PromptScope, server: ServerConnection.Key) { if ("draftID" in scope) return tabs.find((tab) => tab.type === "draft" && tab.draftID === scope.draftID) if (!scope.id) return return ( @@ -189,7 +53,7 @@ export function selectPromptTab(tabs: Tab[], scope: Scope, server: ServerConnect ) } -function scopeKey(scope: Scope) { +function scopeKey(scope: PromptScope) { if ("draftID" in scope) return `draft:${scope.draftID}` return `${scope.dir}:${scope.id ?? WORKSPACE_KEY}` } @@ -199,91 +63,6 @@ type PromptCacheEntry = { dispose: VoidFunction } -function promptTarget(serverScope: ServerScope, scope: Scope) { - if ("draftID" in scope) return Persist.draft(scope.draftID, "prompt") - const legacy = `${scope.dir}/prompt${scope.id ? "/" + scope.id : ""}.v2` - return Persist.serverScoped(serverScope, scope.dir, scope.id, "prompt", [legacy]) -} - -export function createPromptSession(serverScope: ServerScope, scope: Scope) { - const [store, setStore, _, ready] = persisted( - promptTarget(serverScope, scope), - createStore(promptStore()), - ) - - return { ready, ...createPromptStateValue(store, setStore) } -} - -export function createPromptReady(session: Accessor) { - return Object.defineProperty(() => session().ready(), "promise", { - get: () => session().ready.promise, - }) as (() => boolean) & { readonly promise: Promise | undefined } -} - -function promptStore(): PromptStore { - return { - prompt: clonePrompt(DEFAULT_PROMPT), - cursor: undefined, - context: { - items: [], - }, - } -} - -function createPromptStateValue(store: PromptStore, setStore: SetStoreFunction) { - const actions = createPromptActions(setStore) - - const value = { - current: () => store.prompt, - cursor: createMemo(() => store.cursor), - dirty: () => !isPromptEqual(store.prompt, DEFAULT_PROMPT), - context: { - items: createMemo(() => store.context.items), - add(item: ContextItem) { - const key = contextItemKey(item) - if (store.context.items.find((x) => x.key === key)) return - setStore("context", "items", (items) => [...items, { key, ...item }]) - }, - remove(key: string) { - setStore("context", "items", (items) => items.filter((x) => x.key !== key)) - }, - removeComment(path: string, commentID: string) { - setStore("context", "items", (items) => - items.filter((item) => !(item.type === "file" && item.path === path && item.commentID === commentID)), - ) - }, - updateComment(path: string, commentID: string, next: Partial & { comment?: string }) { - setStore("context", "items", (items) => - items.map((item) => { - if (item.type !== "file" || item.path !== path || item.commentID !== commentID) return item - const value = { ...item, ...next } - return { ...value, key: contextItemKey(value) } - }), - ) - }, - replaceComments(items: FileContextItem[]) { - setStore("context", "items", (current) => [ - ...current.filter((item) => !isCommentItem(item)), - ...items.map((item) => ({ ...item, key: contextItemKey(item) })), - ]) - }, - }, - set: actions.set, - reset: actions.reset, - capture: () => value, - } - return value -} - -export function createPromptState() { - const [store, setStore] = createStore(promptStore()) - const ready = Object.assign(() => true, { promise: Promise.resolve(true) }) - return { - ready, - ...createPromptStateValue(store, setStore), - } -} - export const createTabPromptState = ( tabs: ReturnType, tab: Tab, @@ -303,9 +82,7 @@ export const { use: usePrompt, provider: PromptProvider } = createSimpleContext( const cache = new Map() const disposeAll = () => { - for (const entry of cache.values()) { - entry.dispose() - } + for (const entry of cache.values()) entry.dispose() cache.clear() } @@ -324,13 +101,11 @@ export const { use: usePrompt, provider: PromptProvider } = createSimpleContext( const owner = getOwner() const serverKey = () => params.serverKey ? requireServerKey(params.serverKey) : ServerConnection.key(serverSDK().server) - const scope = () => + const scope = (): PromptScope => search.draftId ? { draftID: search.draftId } : { dir: base64Encode(sdk().directory), id: params.id } - const load = (scope: Scope) => { + const load = (scope: PromptScope) => { const current = settings.general.newLayoutDesigns() ? selectPromptTab(tabs.store, scope, serverKey()) : undefined - if (current) { - return createTabPromptState(tabs, current, serverSDK().scope, scope) - } + if (current) return createTabPromptState(tabs, current, serverSDK().scope, scope) const key = scopeKey(scope) const existing = cache.get(key) @@ -354,15 +129,19 @@ export const { use: usePrompt, provider: PromptProvider } = createSimpleContext( } const session = createMemo(() => load(scope())) - const pick = (scope?: Scope) => (scope ? load(scope) : session()) + const pick = (scope?: PromptScope) => (scope ? load(scope) : session()) const ready = createPromptReady(session) return { ready, - capture: (scope?: Scope) => pick(scope).capture(), + capture: (scope?: PromptScope) => pick(scope).capture(), current: () => session().current(), cursor: () => session().cursor(), dirty: () => session().dirty(), + model: { + current: () => session().model.current(), + set: (model: PromptModel | undefined) => session().model.set(model), + }, context: { items: () => session().context.items(), add: (item: ContextItem) => session().context.add(item), @@ -372,8 +151,8 @@ export const { use: usePrompt, provider: PromptProvider } = createSimpleContext( session().context.updateComment(path, commentID, next), replaceComments: (items: FileContextItem[]) => session().context.replaceComments(items), }, - set: (prompt: Prompt, cursorPosition?: number, scope?: Scope) => pick(scope).set(prompt, cursorPosition), - reset: (scope?: Scope) => pick(scope).reset(), + set: (prompt: Prompt, cursorPosition?: number, scope?: PromptScope) => pick(scope).set(prompt, cursorPosition), + reset: (scope?: PromptScope) => pick(scope).reset(), } }, }) diff --git a/packages/app/src/context/tab-memory.ts b/packages/app/src/context/tab-memory.ts index de8050f37971..70bb987c7bb4 100644 --- a/packages/app/src/context/tab-memory.ts +++ b/packages/app/src/context/tab-memory.ts @@ -16,6 +16,9 @@ export function createTabMemory(owner: Owner | null) { } return { + get(key: string, name: string) { + return entries.get(key)?.get(name)?.value as T | undefined + }, ensure(key: string, name: string, init: () => T) { const state = entries.get(key) ?? new Map() if (!entries.has(key)) entries.set(key, state) diff --git a/packages/app/src/context/tabs.test.ts b/packages/app/src/context/tabs.test.ts index b4bff4caff21..b56ea2dfcf93 100644 --- a/packages/app/src/context/tabs.test.ts +++ b/packages/app/src/context/tabs.test.ts @@ -22,6 +22,8 @@ describe("tab memory", () => { }) expect(memory.ensure("tab", "prompt", () => ({ value: "other" }))).toBe(first) + expect(memory.get("tab", "prompt")).toBe(first) + expect(memory.get("missing", "prompt")).toBeUndefined() expect(memory.ensure("other", "prompt", () => ({ value: "other" }))).not.toBe(first) memory.remove("tab") diff --git a/packages/app/src/context/tabs.tsx b/packages/app/src/context/tabs.tsx index 47ed77e781e7..cac6500350a1 100644 --- a/packages/app/src/context/tabs.tsx +++ b/packages/app/src/context/tabs.tsx @@ -11,6 +11,7 @@ import { SessionTabsRemovedDetail } from "@/components/titlebar-session-events" import { sessionHref } from "@/utils/session-route" import { createTabMemory } from "./tab-memory" import { nextTabAfterClose, pushClosedTab, removeClosedTabs, takeClosedTab, type ClosedTab } from "./closed-tabs" +import { createDraftPromptSession, type PromptModel } from "./prompt-state" export type SessionTab = { type: "session" @@ -207,15 +208,17 @@ export const { use: useTabs, provider: TabsProvider } = createSimpleContext({ if (!tab || tab.type !== "draft") throw new Error(`Draft not found: ${draftID}`) return tab }, - newDraft(draft: Omit, prompt?: string) { + newDraft(draft: Omit, prompt?: string, model?: PromptModel) { const draftID = uuid() + const tab = { type: "draft" as const, draftID, ...draft } + memory.ensure(tabKey(tab), "prompt", () => createDraftPromptSession(draftID, { prompt, model })) void startTransition(() => { setStore( produce((tabs) => { - tabs.push({ type: "draft", draftID, ...draft }) + tabs.push(tab) }), ) - navigate(prompt ? `${draftHref(draftID)}&prompt=${encodeURIComponent(prompt)}` : draftHref(draftID)) + navigate(draftHref(draftID)) }) }, updateDraft(draftID: string, draft: Partial>) { @@ -373,6 +376,9 @@ export const { use: useTabs, provider: TabsProvider } = createSimpleContext({ state(tab: Tab, name: string, init: () => T) { return memory.ensure(tabKey(tab), name, init) }, + stateValue(tab: Tab, name: string) { + return memory.get(tabKey(tab), name) + }, } return { ...actions, store, info, ready, recentReady } diff --git a/packages/app/src/i18n/en.ts b/packages/app/src/i18n/en.ts index 9d122447a8d0..1cfdc51782ee 100644 --- a/packages/app/src/i18n/en.ts +++ b/packages/app/src/i18n/en.ts @@ -638,7 +638,7 @@ export const dict = { "session.error.notFound.description": "This tab points to a session that no longer exists on this server.", "session.error.notFound.closeTab": "Close Tab", "session.error.serverConnection": "Can't connect to this server", - "session.review.filesChanged": "{{count}} Files Changed", + "session.review.filesChanged": "Files Changed {{count}}", "session.review.change.one": "Change", "session.review.change.other": "Changes", "session.review.loadingChanges": "Loading changes...", diff --git a/packages/app/src/pages/home-session-open.test.ts b/packages/app/src/pages/home-session-open.test.ts index a71d58ff0723..74b9426654a8 100644 --- a/packages/app/src/pages/home-session-open.test.ts +++ b/packages/app/src/pages/home-session-open.test.ts @@ -2,11 +2,30 @@ import { describe, expect, test } from "bun:test" import { shouldOpenSessionInBackground } from "./home-session-open" describe("shouldOpenSessionInBackground", () => { + test("opens middle clicks in the background", () => { + expect( + shouldOpenSessionInBackground({ button: 1, mac: true, meta: false, ctrl: false, shift: false, alt: false }), + ).toBe(true) + expect( + shouldOpenSessionInBackground({ button: 2, mac: true, meta: false, ctrl: false, shift: false, alt: false }), + ).toBe(false) + }) + test("requires only the platform primary modifier", () => { - expect(shouldOpenSessionInBackground({ mac: true, meta: true, ctrl: false, shift: false, alt: false })).toBe(true) - expect(shouldOpenSessionInBackground({ mac: false, meta: false, ctrl: true, shift: false, alt: false })).toBe(true) - expect(shouldOpenSessionInBackground({ mac: true, meta: true, ctrl: false, shift: true, alt: false })).toBe(false) - expect(shouldOpenSessionInBackground({ mac: false, meta: false, ctrl: true, shift: false, alt: true })).toBe(false) - expect(shouldOpenSessionInBackground({ mac: false, meta: true, ctrl: false, shift: false, alt: false })).toBe(false) + expect( + shouldOpenSessionInBackground({ button: 0, mac: true, meta: true, ctrl: false, shift: false, alt: false }), + ).toBe(true) + expect( + shouldOpenSessionInBackground({ button: 0, mac: false, meta: false, ctrl: true, shift: false, alt: false }), + ).toBe(true) + expect( + shouldOpenSessionInBackground({ button: 0, mac: true, meta: true, ctrl: false, shift: true, alt: false }), + ).toBe(false) + expect( + shouldOpenSessionInBackground({ button: 0, mac: false, meta: false, ctrl: true, shift: false, alt: true }), + ).toBe(false) + expect( + shouldOpenSessionInBackground({ button: 0, mac: false, meta: true, ctrl: false, shift: false, alt: false }), + ).toBe(false) }) }) diff --git a/packages/app/src/pages/home-session-open.ts b/packages/app/src/pages/home-session-open.ts index 8e32efb559d2..9f117935b77e 100644 --- a/packages/app/src/pages/home-session-open.ts +++ b/packages/app/src/pages/home-session-open.ts @@ -1,10 +1,13 @@ export function shouldOpenSessionInBackground(input: { + button: number mac: boolean meta: boolean ctrl: boolean shift: boolean alt: boolean }) { + if (input.button === 1) return true + if (input.button !== 0) return false if (input.shift || input.alt) return false if (input.mac) return input.meta && !input.ctrl return input.ctrl && !input.meta diff --git a/packages/app/src/pages/home.tsx b/packages/app/src/pages/home.tsx index c8a89672dfca..d697e53ae711 100644 --- a/packages/app/src/pages/home.tsx +++ b/packages/app/src/pages/home.tsx @@ -240,10 +240,11 @@ function useHomeSessionHeaderOpacity(groups: () => HomeSessionGroup[]) { return { setViewport, setContentRef, setHeaderRef, update, titleOpacity } } -// Cmd+click on macOS (Ctrl+click elsewhere) opens a session tab in the -// background without navigating, matching browser conventions. +// Middle-click or Cmd+click on macOS (Ctrl+click elsewhere) opens a session +// tab in the background without navigating, matching browser conventions. function isBackgroundOpen(event: MouseEvent) { return shouldOpenSessionInBackground({ + button: event.button, mac: typeof navigator === "object" && /(Mac|iPod|iPhone|iPad)/.test(navigator.platform), meta: event.metaKey, ctrl: event.ctrlKey, @@ -465,8 +466,8 @@ export function NewHome() { } function editProject(conn: ServerConnection.Any, project: LocalProject) { - void import("@/components/dialog-edit-project").then((x) => { - dialog.show(() => ) + void import("@/components/dialog-edit-project-v2").then((x) => { + void dialog.show(() => ) }) } @@ -1386,7 +1387,15 @@ function HomeSessionSearchResultRow(props: { group: !!showProjectName(), }} onMouseEnter={() => props.onHighlight()} + onMouseDown={(event) => { + if (event.button === 1) event.preventDefault() + }} onClick={(event) => props.onSelect(props.record.session, { background: isBackgroundOpen(event) })} + onAuxClick={(event) => { + if (!isBackgroundOpen(event)) return + event.preventDefault() + props.onSelect(props.record.session, { background: true }) + }} > { + if (event.button === 1) event.preventDefault() + }} onClick={(event) => props.openSession(props.record.session, { background: isBackgroundOpen(event) })} + onAuxClick={(event) => { + if (!isBackgroundOpen(event)) return + event.preventDefault() + props.openSession(props.record.session, { background: true }) + }} > sdk().directory) const openProviderSettings = useSettingsDialog("providers") const route = useSessionKey() const [searchParams, setSearchParams] = useSearchParams<{ draftId?: string; prompt?: string }>() + const local = useLocal() + const model = createPromptModelSelection({ agent: local.agent.current }) - useComposerCommands() + useComposerCommands({ model }) let inputRef: HTMLDivElement | undefined @@ -61,6 +67,7 @@ export default function NewSessionPage() { sessionKey: route.sessionKey, sessionID: () => route.params.id, queryOptions: serverSync().queryOptions, + model, }) const projectControls = createPromptProjectControls() const projectController = createPromptProjectController({ @@ -68,6 +75,16 @@ export default function NewSessionPage() { onDone: () => inputRef?.focus(), }) + command.register("new-session", () => [ + { + id: "input.focus", + title: language.t("command.input.focus"), + category: language.t("command.category.view"), + keybind: "ctrl+l", + onSelect: () => inputRef?.focus(), + }, + ]) + const [store, setStore] = createStore<{ worktree?: string }>({}) const rightMount = useTitlebarRightMount() diff --git a/packages/app/src/pages/session.tsx b/packages/app/src/pages/session.tsx index cbf47dc84984..7ce4627135ec 100644 --- a/packages/app/src/pages/session.tsx +++ b/packages/app/src/pages/session.tsx @@ -1,4 +1,5 @@ -import type { Project, UserMessage, VcsFileDiff } from "@opencode-ai/sdk/v2" +import type { FilePart, Project, UserMessage, VcsFileDiff } from "@opencode-ai/sdk/v2" +import { getFilename } from "@opencode-ai/core/util/path" import { useDialog } from "@opencode-ai/ui/context/dialog" import { createQuery, skipToken, useMutation, useQueryClient } from "@tanstack/solid-query" import { @@ -69,7 +70,7 @@ import { MessageTimeline } from "@/pages/session/timeline/message-timeline" import { createTimelineModel } from "@/pages/session/timeline/model" import { type DiffStyle, SessionReviewTab, type SessionReviewTabProps } from "@/pages/session/review-tab" import { useSessionLayout } from "@/pages/session/session-layout" -import { syncSessionModel } from "@/pages/session/session-model-helpers" +import { restorePromptModel, syncPromptModel, syncSessionModel } from "@/pages/session/session-model-helpers" import { clampSessionPanelWidth, SESSION_PANEL_WIDTH_MIN, @@ -79,6 +80,7 @@ import { SessionSidePanel } from "@/pages/session/session-side-panel" import { sessionPanelLayout } from "@/pages/session/session-panel-layout" import { SessionReviewEmptyChangesV2 } from "@opencode-ai/session-ui/v2/session-review-empty-changes-v2" import { SessionReviewEmptyNoGitV2 } from "@opencode-ai/session-ui/v2/session-review-empty-no-git-v2" +import { SessionReviewV2SidebarToggle } from "@opencode-ai/session-ui/v2/session-review-v2" import { ReviewPanelV2 } from "@/pages/session/v2/review-panel-v2" import { createReviewPanelV2State } from "@/pages/session/v2/review-panel-v2-state" import { reviewDiffDirectory, reviewDiffNeedsLoad, reviewRootDirectory } from "@/pages/session/v2/review-diff-kinds" @@ -483,7 +485,7 @@ export default function Page() { if (desktopSessionResizeOpen()) return `${sessionPanelResizedWidth()}px` return `calc(100% - ${layout.fileTree.width()}px)` }) - const centered = createMemo(() => isDesktop() && !desktopReviewOpen()) + const centered = createMemo(() => isDesktop() && (newSessionDesign() || !desktopReviewOpen())) const desktopV2PanelLayout = createMemo(() => sessionPanelLayout({ review: desktopV2ReviewOpen(), @@ -557,6 +559,17 @@ export default function Page() { ), ) + let restoredModelSession: string | undefined + createEffect(() => { + const id = params.id + if (!id || !prompt.ready() || !local.session.ready()) return + if (restoredModelSession !== id) { + restoredModelSession = id + if (restorePromptModel(local, prompt)) return + } + syncPromptModel(local, prompt) + }) + createEffect( on( () => ({ dir: sdk().directory, id: params.id }), @@ -1304,7 +1317,7 @@ export default function Page() { const reviewPanelV2Rendered = createMemo((prev) => prev || !store.deferRender, false) const reviewPanelV2 = () => ( -
+
@@ -1893,7 +1906,30 @@ export default function Page() { .map((item) => ({ id: item.id, text: line(item.id) })) }) - const actions = { revert } + // attachment bytes are embedded as a data URL, so downloading always works; + // revealing requires the on-disk path captured by the client that attached the file + const openAttachment = (file: FilePart) => { + const download = () => { + const anchor = document.createElement("a") + anchor.href = file.url + anchor.download = getFilename(file.filename) || "attachment" + anchor.click() + } + const path = file.filename ?? "" + const absolute = path.startsWith("/") || path.startsWith("\\\\") || /^[a-zA-Z]:[\\/]/.test(path) + if (platform.revealPath && absolute) { + void platform.revealPath(path).then( + (revealed) => { + if (!revealed) download() + }, + () => download(), + ) + return + } + download() + } + + const actions = { revert, openAttachment } createEffect(() => { const sessionID = params.id @@ -2265,6 +2301,13 @@ export default function Page() { reviewHasFocusableContent={() => hasReview() || reviewV2State.sidebarOpened()} reviewCount={reviewCount} reviewPanel={reviewPanelV2} + reviewSidebarToggle={(disabled) => ( + + )} fileBrowserState={reviewV2State} activeDiff={activeReviewFile()} focusReviewDiff={focusReviewDiff} diff --git a/packages/app/src/pages/session/composer/prompt-model-selection.ts b/packages/app/src/pages/session/composer/prompt-model-selection.ts new file mode 100644 index 000000000000..3f5bb0f477e8 --- /dev/null +++ b/packages/app/src/pages/session/composer/prompt-model-selection.ts @@ -0,0 +1,133 @@ +import { batch, createMemo, startTransition } from "solid-js" +import { useModels } from "@/context/models" +import type { ModelKey, ModelSelection } from "@/context/local" +import { cycleModelVariant, getConfiguredAgentVariant, resolveModelVariant } from "@/context/model-variant" +import { usePrompt } from "@/context/prompt" +import { useSDK } from "@/context/sdk" +import { useSync } from "@/context/sync" +import { useProviders } from "@/hooks/use-providers" + +export function createPromptModelSelection(input: { agent: () => { model?: ModelKey; variant?: string } | undefined }) { + const sdk = useSDK() + const sync = useSync() + const models = useModels() + const prompt = usePrompt() + const providers = useProviders(() => sdk().directory) + const connected = createMemo(() => new Set(providers.connected().map((item) => item.id))) + + const valid = (model: ModelKey) => { + const provider = providers.all().get(model.providerID) + return !!provider?.models[model.modelID] && connected().has(model.providerID) + } + + const configured = () => { + const value = sync().data.config.model + if (!value) return + const [providerID, modelID] = value.split("/") + const model = { providerID, modelID } + if (valid(model)) return model + } + + const recent = () => models.recent.list().find(valid) + const fallback = () => { + const defaults = providers.default() + return providers.connected().flatMap((provider) => { + const modelID = defaults[provider.id] ?? Object.values(provider.models)[0]?.id + return modelID ? [{ providerID: provider.id, modelID }] : [] + })[0] + } + + const current = () => { + const key = [prompt.model.current(), input.agent()?.model, configured(), recent(), fallback()].find( + (item): item is ModelKey => !!item && valid(item), + ) + if (!key) return + return models.find(key) + } + const recentModels = createMemo(() => + models.recent + .list() + .map(models.find) + .filter((item): item is NonNullable => !!item), + ) + + const selection = { + ready: models.ready, + current, + recent: recentModels, + list: models.list, + cycle(direction: 1 | -1) { + const items = recentModels() + const item = current() + if (!item) return + const index = items.findIndex((entry) => entry.provider.id === item.provider.id && entry.id === item.id) + if (index === -1) return + const next = items[(index + direction + items.length) % items.length] + if (next) selection.set({ providerID: next.provider.id, modelID: next.id }) + }, + set(item: ModelKey | undefined, options?: { recent?: boolean }) { + startTransition(() => + batch(() => { + prompt.model.set(item ? { ...item, variant: prompt.model.current()?.variant } : undefined) + if (!item) return + models.setVisibility(item, true) + if (options?.recent) models.recent.push(item) + }), + ) + }, + visible: models.visible, + setVisibility: models.setVisibility, + variant: { + configured() { + const item = input.agent() + const model = current() + if (!item || !model) return + return getConfiguredAgentVariant({ + agent: { model: item.model, variant: item.variant }, + model: { providerID: model.provider.id, modelID: model.id, variants: model.variants }, + }) + }, + selected() { + return prompt.model.current()?.variant + }, + current() { + const resolved = resolveModelVariant({ + variants: this.list(), + selected: this.selected(), + configured: this.configured(), + }) + if (resolved) return resolved + const model = current() + if (!model) return + const saved = models.variant.get({ providerID: model.provider.id, modelID: model.id }) + if (saved && this.list().includes(saved)) return saved + }, + list() { + return Object.keys(current()?.variants ?? {}) + }, + set(value: string | undefined) { + startTransition(() => + batch(() => { + const model = current() + if (!model) return + prompt.model.set({ providerID: model.provider.id, modelID: model.id, variant: value ?? null }) + models.variant.set({ providerID: model.provider.id, modelID: model.id }, value) + }), + ) + }, + cycle() { + const variants = this.list() + if (variants.length === 0) return + this.set( + cycleModelVariant({ + variants, + selected: this.selected(), + configured: this.configured(), + }), + ) + }, + }, + } satisfies ModelSelection + + return selection +} diff --git a/packages/app/src/pages/session/composer/session-composer-controls.ts b/packages/app/src/pages/session/composer/session-composer-controls.ts index c7f2304531a5..8d5d96719c06 100644 --- a/packages/app/src/pages/session/composer/session-composer-controls.ts +++ b/packages/app/src/pages/session/composer/session-composer-controls.ts @@ -7,7 +7,7 @@ import type { PromptProjectControls } from "@/components/prompt-project-selector import { useDirectoryPicker } from "@/components/directory-picker" import { useGlobal } from "@/context/global" import { useLayout } from "@/context/layout" -import { useLocal } from "@/context/local" +import { useLocal, type ModelSelection } from "@/context/local" import type { QueryOptionsApi } from "@/context/server-sync" import { useServerSDK } from "@/context/server-sdk" import { serverName, ServerConnection, useServer } from "@/context/server" @@ -22,6 +22,7 @@ export function createPromptInputController(input: { sessionKey: Accessor sessionID: Accessor queryOptions: Pick + model?: ModelSelection }) { const layout = useLayout() const local = useLocal() @@ -44,7 +45,7 @@ export function createPromptInputController(input: { select: local.agent.set, }, model: { - selection: local.model, + selection: input.model ?? local.model, paid: providers.paid().length > 0, loading: agentsQuery.isLoading || providersQuery.isLoading || globalProvidersQuery.isLoading, }, diff --git a/packages/app/src/pages/session/file-tabs.tsx b/packages/app/src/pages/session/file-tabs.tsx index dd51dd05852e..b54fbc4e4c48 100644 --- a/packages/app/src/pages/session/file-tabs.tsx +++ b/packages/app/src/pages/session/file-tabs.tsx @@ -1,4 +1,4 @@ -import { createEffect, createMemo, createSignal, Match, on, onCleanup, Switch } from "solid-js" +import { createEffect, createMemo, createSignal, Match, on, onCleanup, Show, Switch } from "solid-js" import { createStore } from "solid-js/store" import { Dynamic } from "solid-js/web" import { makeEventListener } from "@solid-primitives/event-listener" @@ -6,9 +6,12 @@ import type { FileSearchHandle } from "@opencode-ai/session-ui/file" import { useFileComponent } from "@opencode-ai/ui/context/file" import { cloneSelectedLineRange, previewSelectedLines } from "@opencode-ai/session-ui/pierre/selection-bridge" import { createLineCommentController } from "@opencode-ai/session-ui/line-comment-annotations" +import { createLineCommentControllerV2 } from "@opencode-ai/session-ui/v2/line-comment-annotations-v2" import { sampledChecksum } from "@opencode-ai/core/util/encode" import { DropdownMenu } from "@opencode-ai/ui/dropdown-menu" import { IconButton } from "@opencode-ai/ui/icon-button" +import { LineCommentV2OverflowIcon } from "@opencode-ai/ui/v2/line-comment-v2" +import { MenuV2 } from "@opencode-ai/ui/v2/menu-v2" import { Tabs } from "@opencode-ai/ui/tabs" import { ScrollView } from "@opencode-ai/ui/scroll-view" import { showToast } from "@/utils/toast" @@ -16,6 +19,7 @@ import { selectionFromLines, useFile, type FileSelection, type SelectedLineRange import { useComments } from "@/context/comments" import { useLanguage } from "@/context/language" import { usePrompt } from "@/context/prompt" +import { useSettings } from "@/context/settings" import { getSessionHandoff } from "@/pages/session/handoff" import { useSessionLayout } from "@/pages/session/session-layout" import { createSessionTabs } from "@/pages/session/helpers" @@ -53,6 +57,30 @@ function FileCommentMenu(props: { ) } +function FileCommentMenuV2(props: { + moreLabel: string + editLabel: string + deleteLabel: string + onEdit: VoidFunction + onDelete: VoidFunction +}) { + return ( +
event.stopPropagation()} onClick={(event) => event.stopPropagation()}> + + + + + + + {props.editLabel} + {props.deleteLabel} + + + +
+ ) +} + type ScrollPos = { x: number; y: number } function createScrollSync(input: { tab: () => string; view: ReturnType["view"] }) { @@ -180,6 +208,15 @@ export function FileTabContent(props: { tab: string }) { } export function SessionFileView(props: { tab: string }) { + const settings = useSettings() + return ( + }> + + + ) +} + +function SessionFileViewV1(props: { tab: string }) { const file = useFile() const comments = useComments() const language = useLanguage() @@ -463,3 +500,294 @@ export function SessionFileView(props: { tab: string }) { return content() } + +function SessionFileViewV2(props: { tab: string }) { + const file = useFile() + const comments = useComments() + const language = useLanguage() + const prompt = usePrompt() + const fileComponent = useFileComponent() + const { sessionKey, tabs, view } = useSessionLayout() + const activeFileTab = createSessionTabs({ + tabs, + pathFromTab: file.pathFromTab, + normalizeTab: (tab) => (tab.startsWith("file://") ? file.tab(tab) : tab), + }).activeFileTab + + let find: FileSearchHandle | null = null + + const search = { + register: (handle: FileSearchHandle | null) => { + find = handle + }, + } + + const path = createMemo(() => file.pathFromTab(props.tab)) + const state = createMemo(() => { + const p = path() + if (!p) return + return file.get(p) + }) + const contents = createMemo(() => state()?.content?.content ?? "") + const cacheKey = createMemo(() => sampledChecksum(contents())) + const selectedLines = createMemo(() => { + const p = path() + if (!p) return null + if (file.ready()) return (file.selectedLines(p) as SelectedLineRange | undefined) ?? null + return (getSessionHandoff(sessionKey())?.files[p] as SelectedLineRange | undefined) ?? null + }) + const scrollSync = createScrollSync({ + tab: () => props.tab, + view, + }) + + const selectionPreview = (source: string, selection: FileSelection) => { + return previewSelectedLines(source, { + start: selection.startLine, + end: selection.endLine, + }) + } + + const buildPreview = (filePath: string, selection: FileSelection) => { + const source = filePath === path() ? contents() : file.get(filePath)?.content?.content + if (!source) return undefined + return selectionPreview(source, selection) + } + + const addCommentToContext = (input: { + file: string + selection: SelectedLineRange + comment: string + preview?: string + origin?: "review" | "file" + }) => { + const selection = selectionFromLines(input.selection) + const preview = input.preview ?? buildPreview(input.file, selection) + + const saved = comments.add({ + file: input.file, + selection: input.selection, + comment: input.comment, + }) + prompt.context.add({ + type: "file", + path: input.file, + selection, + comment: input.comment, + commentID: saved.id, + commentOrigin: input.origin, + preview, + }) + } + + const updateCommentInContext = (input: { + id: string + file: string + selection: SelectedLineRange + comment: string + }) => { + comments.update(input.file, input.id, input.comment) + const preview = input.file === path() ? buildPreview(input.file, selectionFromLines(input.selection)) : undefined + prompt.context.updateComment(input.file, input.id, { + comment: input.comment, + ...(preview ? { preview } : {}), + }) + } + + const removeCommentFromContext = (input: { id: string; file: string }) => { + comments.remove(input.file, input.id) + prompt.context.removeComment(input.file, input.id) + } + + const fileComments = createMemo(() => { + const p = path() + if (!p) return [] + return comments.list(p) + }) + + const commentedLines = createMemo(() => fileComments().map((comment) => comment.selection)) + + const [note, setNote] = createStore({ + openedComment: null as string | null, + commenting: null as SelectedLineRange | null, + selected: null as SelectedLineRange | null, + }) + + const syncSelected = (range: SelectedLineRange | null) => { + const p = path() + if (!p) return + file.setSelectedLines(p, range ? cloneSelectedLineRange(range) : null) + } + + const activeSelection = () => note.selected ?? selectedLines() + + const commentsUi = createLineCommentControllerV2({ + comments: fileComments, + label: language.t("ui.lineComment.submit"), + draftKey: () => path() ?? props.tab, + mention: { + items: file.searchFilesAndDirectories, + }, + getSide: (range) => range.endSide ?? range.side ?? "additions", + state: { + opened: () => note.openedComment, + setOpened: (id) => setNote("openedComment", id), + selected: () => note.selected, + setSelected: (range) => setNote("selected", range), + commenting: () => note.commenting, + setCommenting: (range) => setNote("commenting", range), + syncSelected, + hoverSelected: syncSelected, + }, + onSubmit: ({ comment, selection }) => { + const p = path() + if (!p) return + addCommentToContext({ file: p, selection, comment, origin: "file" }) + }, + onUpdate: ({ id, comment, selection }) => { + const p = path() + if (!p) return + updateCommentInContext({ id, file: p, selection, comment }) + }, + onDelete: (comment) => { + const p = path() + if (!p) return + removeCommentFromContext({ id: comment.id, file: p }) + }, + editSubmitLabel: language.t("common.save"), + renderCommentActions: (_, controls) => ( + + ), + }) + + createEffect(() => { + if (typeof window === "undefined") return + + const onKeyDown = (event: KeyboardEvent) => { + if (activeFileTab() !== props.tab) return + if (!(event.metaKey || event.ctrlKey) || event.altKey || event.shiftKey) return + if (event.key.toLowerCase() !== "f") return + + event.preventDefault() + event.stopPropagation() + find?.focus() + } + + makeEventListener(window, "keydown", onKeyDown, { capture: true }) + }) + + createEffect( + on( + path, + () => { + commentsUi.note.reset() + }, + { defer: true }, + ), + ) + + createEffect(() => { + const focus = comments.focus() + const p = path() + if (!focus || !p) return + if (focus.file !== p) return + if (activeFileTab() !== props.tab) return + + const target = fileComments().find((comment) => comment.id === focus.id) + if (!target) return + + commentsUi.note.openComment(target.id, target.selection, { cancelDraft: true }) + requestAnimationFrame(() => comments.clearFocus()) + }) + + let prev = { + loaded: false, + ready: false, + active: false, + } + + createEffect(() => { + const loaded = !!state()?.loaded + const ready = file.ready() + const active = activeFileTab() === props.tab + const restore = (loaded && !prev.loaded) || (ready && !prev.ready) || (active && loaded && !prev.active) + prev = { loaded, ready, active } + if (!restore) return + scrollSync.queueRestore() + }) + + const renderFile = (source: string) => ( +
+ { + scrollSync.queueRestore() + }} + annotations={commentsUi.annotations()} + renderAnnotation={commentsUi.renderAnnotation} + renderGutterUtility={commentsUi.renderGutterUtility} + onLineSelected={(range: SelectedLineRange | null) => { + commentsUi.onLineSelected(range) + }} + onLineSelectionEnd={(range: SelectedLineRange | null) => { + if (!range) { + commentsUi.note.select(null) + commentsUi.note.cancelDraft() + return + } + commentsUi.onLineSelectionEnd(range) + }} + onLineNumberSelectionEnd={(range: SelectedLineRange | null) => { + commentsUi.onLineNumberSelectionEnd(range) + }} + search={search} + class="select-text" + media={{ + mode: "auto", + path: path(), + current: state()?.content, + onLoad: scrollSync.queueRestore, + onError: (args: { kind: "image" | "audio" | "svg" }) => { + if (args.kind !== "svg") return + showToast({ + variant: "error", + title: language.t("toast.file.loadFailed.title"), + }) + }, + }} + /> +
+ ) + + const content = () => ( +
+ + + {renderFile(contents())} + +
{language.t("common.loading")}...
+
+ {(err) =>
{err()}
}
+
+
+
+ ) + + return content() +} diff --git a/packages/app/src/pages/session/session-model-helpers.test.ts b/packages/app/src/pages/session/session-model-helpers.test.ts index 2ab293b8fb3a..201627b3691e 100644 --- a/packages/app/src/pages/session/session-model-helpers.test.ts +++ b/packages/app/src/pages/session/session-model-helpers.test.ts @@ -1,6 +1,6 @@ import { describe, expect, test } from "bun:test" import type { UserMessage } from "@opencode-ai/sdk/v2" -import { resetSessionModel, syncSessionModel } from "./session-model-helpers" +import { resetSessionModel, restorePromptModel, syncPromptModel, syncSessionModel } from "./session-model-helpers" const message = (input?: { agent?: string; model?: UserMessage["model"] }) => ({ @@ -50,3 +50,102 @@ describe("resetSessionModel", () => { expect(calls).toEqual(["reset"]) }) }) + +describe("syncPromptModel", () => { + test("stores the effective session model in prompt state", () => { + const calls: unknown[] = [] + + syncPromptModel( + { + model: { + current: () => ({ id: "claude-sonnet-4", provider: { id: "anthropic" } }), + set() {}, + variant: { current: () => "high", set() {} }, + }, + }, + { + model: { + current: () => undefined, + set: (model) => calls.push(model), + }, + }, + ) + + expect(calls).toEqual([{ providerID: "anthropic", modelID: "claude-sonnet-4", variant: "high" }]) + }) + + test("does not rewrite an unchanged prompt model", () => { + const calls: unknown[] = [] + const model = { providerID: "anthropic", modelID: "claude-sonnet-4", variant: "high" } + + syncPromptModel( + { + model: { + current: () => ({ id: model.modelID, provider: { id: model.providerID } }), + set() {}, + variant: { current: () => model.variant, set() {} }, + }, + }, + { + model: { + current: () => model, + set: (value) => calls.push(value), + }, + }, + ) + + expect(calls).toEqual([]) + }) +}) + +describe("restorePromptModel", () => { + test("restores the persisted prompt model into session selection", () => { + const calls: unknown[] = [] + const restored = restorePromptModel( + { + model: { + current: () => ({ id: "gpt", provider: { id: "openai" } }), + set: (model) => calls.push(model), + variant: { + current: () => undefined, + set: (variant) => calls.push(variant), + }, + }, + }, + { + model: { + current: () => ({ providerID: "anthropic", modelID: "claude", variant: "high" }), + set() {}, + }, + }, + ) + + expect(restored).toBe(true) + expect(calls).toEqual([{ providerID: "anthropic", modelID: "claude" }, "high"]) + }) + + test("does nothing without a persisted prompt model", () => { + const calls: unknown[] = [] + const restored = restorePromptModel( + { + model: { + current: () => ({ id: "gpt", provider: { id: "openai" } }), + set: (model) => calls.push(model), + variant: { + current: () => undefined, + set: (variant) => calls.push(variant), + }, + }, + }, + { + model: { + current: () => undefined, + set() {}, + }, + }, + ) + + expect(restored).toBe(false) + expect(calls).toEqual([]) + }) +}) diff --git a/packages/app/src/pages/session/session-model-helpers.ts b/packages/app/src/pages/session/session-model-helpers.ts index c9e2e1dbd2ae..f2aadff27d84 100644 --- a/packages/app/src/pages/session/session-model-helpers.ts +++ b/packages/app/src/pages/session/session-model-helpers.ts @@ -7,6 +7,24 @@ type Local = { } } +type ModelSelection = { + model: { + current(): { id: string; provider: { id: string } } | undefined + set(model: { providerID: string; modelID: string }): void + variant: { + current(): string | undefined + set(variant: string | undefined): void + } + } +} + +type PromptState = { + model: { + current(): { providerID: string; modelID: string; variant?: string | null } | undefined + set(model: { providerID: string; modelID: string; variant?: string | null }): void + } +} + export const resetSessionModel = (local: Local) => { local.session.reset() } @@ -14,3 +32,32 @@ export const resetSessionModel = (local: Local) => { export const syncSessionModel = (local: Local, msg: UserMessage) => { local.session.restore(msg) } + +export const syncPromptModel = (local: ModelSelection, prompt: PromptState) => { + const model = local.model.current() + if (!model) return + const next = { + providerID: model.provider.id, + modelID: model.id, + variant: local.model.variant.current(), + } + const current = prompt.model.current() + if (current?.providerID === next.providerID && current.modelID === next.modelID && current.variant === next.variant) + return + prompt.model.set(next) +} + +export const restorePromptModel = (local: ModelSelection, prompt: PromptState) => { + const model = prompt.model.current() + if (!model) return false + const current = local.model.current() + if ( + current?.provider.id === model.providerID && + current.id === model.modelID && + local.model.variant.current() === (model.variant ?? undefined) + ) + return true + local.model.set({ providerID: model.providerID, modelID: model.modelID }) + local.model.variant.set(model.variant ?? undefined) + return true +} diff --git a/packages/app/src/pages/session/session-side-panel.tsx b/packages/app/src/pages/session/session-side-panel.tsx index 5e5f89e3e220..184fa684ecc5 100644 --- a/packages/app/src/pages/session/session-side-panel.tsx +++ b/packages/app/src/pages/session/session-side-panel.tsx @@ -1,28 +1,46 @@ import { For, Match, Show, Switch, createEffect, createMemo, onCleanup, type JSX } from "solid-js" import { createStore } from "solid-js/store" import { createMediaQuery } from "@solid-primitives/media" +import { DragDropProvider as DndKitProvider, PointerSensor } from "@dnd-kit/solid" +import { isSortable } from "@dnd-kit/solid/sortable" +import { Accessibility, AutoScroller, Feedback, PointerActivationConstraints } from "@dnd-kit/dom" +import { RestrictToHorizontalAxis } from "@dnd-kit/abstract/modifiers" +import { RestrictToElement } from "@dnd-kit/dom/modifiers" +import { + DragDropProvider, + DragDropSensors, + DragOverlay, + SortableProvider, + closestCenter, + type DragEvent, +} from "@thisbeyond/solid-dnd" import { Tabs } from "@opencode-ai/ui/tabs" import { IconButton } from "@opencode-ai/ui/icon-button" import { Icon } from "@opencode-ai/ui/icon" import { TooltipKeybind } from "@opencode-ai/ui/tooltip" import { ResizeHandle } from "@opencode-ai/ui/resize-handle" import { Mark } from "@opencode-ai/ui/logo" -import { DragDropProvider, DragDropSensors, DragOverlay, SortableProvider, closestCenter } from "@thisbeyond/solid-dnd" -import type { DragEvent } from "@thisbeyond/solid-dnd" +import { IconButtonV2 } from "@opencode-ai/ui/v2/icon-button-v2" +import { KeybindV2 } from "@opencode-ai/ui/v2/keybind-v2" +import { TooltipV2 } from "@opencode-ai/ui/v2/tooltip-v2" import type { SnapshotFileDiff, VcsFileDiff } from "@opencode-ai/sdk/v2" import { ConstrainDragYAxis, getDraggableId } from "@/utils/solid-dnd" import { useDialog } from "@opencode-ai/ui/context/dialog" import FileTree from "@/components/file-tree" +import { normalizeFileTreeV2Path } from "@/components/file-tree-v2-model" import { SessionContextUsage } from "@/components/session-context-usage" const reviewTabID = "session-side-panel-review-tab" const reviewTabPanelID = "session-side-panel-review-tabpanel" -import { SessionContextTab, SortableTab, FileVisual } from "@/components/session" +const fileBrowserTabPanelID = "session-side-panel-file-browser-tabpanel" +import { SessionContextTab, SortableTab, SortableTabV2, FileVisual } from "@/components/session" +import { OpenInAppV2 } from "@/components/session/open-in-app-v2" import { useCommand } from "@/context/command" import { useFile, type SelectedLineRange } from "@/context/file" import { useLanguage } from "@/context/language" import { useLayout } from "@/context/layout" +import { useSDK } from "@/context/sdk" import { useSettings } from "@/context/settings" import { createFileTabListSync } from "@/pages/session/file-tab-scroll" import { FileTabContent } from "@/pages/session/file-tabs" @@ -53,6 +71,7 @@ export function SessionSidePanel(props: { reviewHasFocusableContent: () => boolean reviewCount: () => number reviewPanel: () => JSX.Element + reviewSidebarToggle?: (disabled: boolean) => JSX.Element fileBrowserState?: SessionFileBrowserState activeDiff?: string focusReviewDiff: (path: string) => void @@ -66,7 +85,9 @@ export function SessionSidePanel(props: { const language = useLanguage() const command = useCommand() const dialog = useDialog() + const sdk = useSDK() const { sessionKey, tabs, view, params } = useSessionLayout() + const projectDirectory = createMemo(() => sdk().directory) const isDesktop = createMediaQuery("(min-width: 768px)") const shown = settings.visibility.fileTree @@ -98,11 +119,9 @@ export function SessionSidePanel(props: { return "mix" as const } - const normalize = (p: string) => p.replaceAll("\\\\", "/").replace(/\/+$/, "") - const out = new Map() for (const diff of diffs()) { - const file = normalize(diff.file) + const file = normalizeFileTreeV2Path(diff.file) const kind = diff.status === "added" ? "add" : diff.status === "deleted" ? "del" : "mix" out.set(file, kind) @@ -159,6 +178,7 @@ export function SessionSidePanel(props: { fileBrowser: () => !!props.fileBrowserState, }) const contextOpen = tabState.contextOpen + const openFileOpen = tabState.openFileOpen const panelTabs = tabState.panelTabs const openedTabs = tabState.openedTabs const activeTab = tabState.activeTab @@ -176,10 +196,8 @@ export function SessionSidePanel(props: { layout.fileTree.setTab("all") } - const [store, setStore] = createStore({ - activeDraggable: undefined as string | undefined, - }) let fileFilter: HTMLInputElement | undefined + let tabList: HTMLDivElement | undefined const temporaryTab = tabs().preview const previewTab = (value: string) => { const next = normalizeTab(value) @@ -202,10 +220,26 @@ export function SessionSidePanel(props: { } const browserTab = createMemo(() => { if (!props.fileBrowserState) return undefined - if (activeTab() === SESSION_OPEN_FILE_TAB) return SESSION_OPEN_FILE_TAB + const active = activeTab() + if (active === SESSION_OPEN_FILE_TAB) return SESSION_OPEN_FILE_TAB + if (active && file.pathFromTab(active)) return active return activeFileTab() }) - const browserKinds = createMemo(() => new Map([...kinds()].filter(([, kind]) => kind !== "mix"))) + // Keep the file-browser shell mounted while any file tab exists. Kobalte briefly + // selects Review while the tab For replaces a preview trigger, which would + // otherwise dispose the sidebar and reset scroll. + const fileBrowserMounted = createMemo(() => { + if (!props.fileBrowserState) return false + return openedTabs().length > 0 || openFileOpen() || !!browserTab() + }) + const fileBrowserVisible = createMemo(() => { + const active = activeTab() + return active !== "review" && active !== "context" && active !== "empty" + }) + const openFileKeybind = createMemo(() => command.keybindParts("file.open")) + const [store, setStore] = createStore({ + activeDraggable: undefined as string | undefined, + }) const handleDragStart = (event: unknown) => { const id = getDraggableId(event) @@ -255,8 +289,10 @@ export function SessionSidePanel(props: { aria-label={language.t("session.panel.reviewAndFiles")} aria-hidden={!open()} inert={!open()} - class="relative min-w-0 flex overflow-hidden bg-background-base" + class="relative min-w-0 flex overflow-hidden" classList={{ + "bg-v2-background-bg-base": settings.general.newLayoutDesigns(), + "bg-background-base": !settings.general.newLayoutDesigns(), "h-full shrink-0": !props.stacked, "h-full min-h-0": props.stacked, "pointer-events-none": !open(), @@ -275,74 +311,297 @@ export function SessionSidePanel(props: { }} > -
-
- - - - -
- { - const stop = createFileTabListSync({ el, contextOpen }) - onCleanup(stop) - }} - > - - +
+ + + + +
+ { + const stop = createFileTabListSync({ el, contextOpen }) + onCleanup(stop) + }} > -
-
{language.t("session.tab.review")}
- -
{props.reviewCount()}
-
-
- - - - + +
+
{language.t("session.tab.review")}
+ +
{props.reviewCount()}
+
+
+
+
+ + + tabs().close("context")} + aria-label={language.t("common.closeTab")} + /> + + } + hideCloseButton + onMiddleClick={() => tabs().close("context")} + > +
+ +
{language.t("session.tab.context")}
+
+
+
+ + + {(tab) => ( + + } + > + + tabs().close(SESSION_OPEN_FILE_TAB)} + aria-label={language.t("common.closeTab")} + /> + + } + hideCloseButton + onMiddleClick={() => tabs().close(SESSION_OPEN_FILE_TAB)} + > +
+ + {language.t("command.file.open")} +
+
+
+ )} +
+
+
tabs().close("context")} - aria-label={language.t("common.closeTab")} + iconSize="large" + class="!rounded-md" + onClick={() => { + void import("@/components/dialog-select-file").then((x) => { + dialog.show(() => ) + }) + }} + aria-label={language.t("command.file.open")} /> - } - hideCloseButton - onMiddleClick={() => tabs().close("context")} +
+
+
+ + +
-
- -
{language.t("session.tab.context")}
+ {props.reviewPanel()} +
+ + + + +
+
+ +
+ {language.t("session.files.selectToOpen")} +
+
- +
- + + + +
+ +
+
+
+ + + {(tab) => } + + + + + {(tab) => { + const path = file.pathFromTab(tab) + return ( +
+ + {(p) => } + +
+ ) + }} +
+
+ + } + > + + event.target instanceof Element && + (!!event.target.closest('[data-slot="tabs-trigger-close-button"]') || + !!event.target.closest(".session-review-v2-open-in-app-slot")), + }), + ]} + modifiers={[ + RestrictToHorizontalAxis, + RestrictToElement.configure({ element: () => tabList ?? null }), + ]} + plugins={(defaults) => [ + ...defaults.filter((plugin) => plugin !== Accessibility), + AutoScroller.configure({ acceleration: 8, threshold: { x: 0.05, y: 0 } }), + Feedback.configure({ dropAnimation: null }), + ]} + onDragEnd={(event) => { + const source = event.operation.source + if (event.canceled || !isSortable(source) || source.initialIndex === source.index) return + tabs().move(source.id.toString(), source.index) + }} + > + +
+ { + tabList = el + const stop = createFileTabListSync({ el, contextOpen }) + onCleanup(stop) + }} + > + + {(toggle) => ( +
+ {toggle()(activeTab() === SESSION_OPEN_FILE_TAB)} +
+ )} +
+ + + {props.hasReview() + ? language.t("session.review.filesChanged", { count: props.reviewCount() }) + : language.t("session.tab.review")} + + + + + tabs().close("context")} + aria-label={language.t("common.closeTab")} + /> + + } + hideCloseButton + onMiddleClick={() => tabs().close("context")} + > +
+ +
{language.t("session.tab.context")}
+
+
+
{(tab) => ( tabs().all().indexOf(tab)} temporary={temporaryTab() === tab} onTabClose={tabs().close} onTabDoubleClick={temporaryTab() === tab ? openTab : undefined} @@ -378,100 +637,104 @@ export function SessionSidePanel(props: { )} - -
- - { - if (props.fileBrowserState) { - openFileBrowser() - return - } - void import("@/components/dialog-select-file").then((x) => { - dialog.show(() => ) - }) - }} - aria-label={language.t("command.file.open")} - /> - + + {language.t("command.file.open")} + 0}> + + + + } + placement="bottom" + class="flex items-center" + > + } + variant="ghost-muted" + size="large" + onClick={() => openFileBrowser()} + aria-label={language.t("command.file.open")} + /> + +
+
+
event.stopPropagation()} + onClick={(event) => event.stopPropagation()} + > +
- -
- - -
- {props.reviewPanel()}
-
- - - -
-
- -
- {language.t("session.files.selectToOpen")} + + +
+ {props.reviewPanel()} +
+
+ + + +
+
+ +
+ {language.t("session.files.selectToOpen")} +
-
- - + + - - -
- -
-
-
- - - previewTab(file.tab(path))} - onSelectPermanent={(path) => openTab(file.tab(path))} - filterRef={(element) => (fileFilter = element)} - /> - - - - {(tab) => } - - - - - {(tab) => { - const path = file.pathFromTab(tab) - return ( -
- - {(p) => } - + + +
+
- ) - }} -
- - + + + + +
+ previewTab(file.tab(path))} + onSelectPermanent={(path) => openTab(file.tab(path))} + filterRef={(element) => (fileFilter = element)} + /> +
+
+ + +
@@ -499,10 +762,19 @@ export function SessionSidePanel(props: { > - {props.reviewCount()}{" "} - {language.t( - props.reviewCount() === 1 ? "session.review.change.one" : "session.review.change.other", - )} + + {props.reviewCount()}{" "} + {language.t( + props.reviewCount() === 1 ? "session.review.change.one" : "session.review.change.other", + )} + + } + > + {language.t("session.review.filesChanged", { count: props.reviewCount() })} + {language.t("session.files.all")} diff --git a/packages/app/src/pages/session/terminal-panel-v2.tsx b/packages/app/src/pages/session/terminal-panel-v2.tsx index e59513cde779..b74d0fe227c7 100644 --- a/packages/app/src/pages/session/terminal-panel-v2.tsx +++ b/packages/app/src/pages/session/terminal-panel-v2.tsx @@ -195,7 +195,7 @@ export function TerminalPanelV2(props: { stacked?: boolean } = {}) { aria-label={language.t("terminal.title")} aria-hidden={!opened()} inert={!opened()} - class="relative shrink-0 overflow-hidden bg-background-stronger" + class="relative shrink-0 overflow-hidden bg-v2-background-bg-base" classList={{ "w-full": !isDesktop() || stacked(), "min-w-0 h-full flex-1": isDesktop() && opened() && !stacked(), @@ -237,7 +237,7 @@ export function TerminalPanelV2(props: { stacked?: boolean } = {}) { when={terminal.ready()} fallback={
-
+
{(title) => (
diff --git a/packages/app/src/pages/session/timeline/message-timeline.tsx b/packages/app/src/pages/session/timeline/message-timeline.tsx index d6cc7be972d6..8a1e9c5fae40 100644 --- a/packages/app/src/pages/session/timeline/message-timeline.tsx +++ b/packages/app/src/pages/session/timeline/message-timeline.tsx @@ -327,6 +327,7 @@ export function MessageTimeline(props: { parts: getMsgParts, status: sessionStatus, showReasoningSummaries: settings.general.showReasoningSummaries, + inlineComments: settings.general.newLayoutDesigns, }) const activeMessageID = projection.activeMessageID const assistantMessagesByParent = projection.assistantMessagesByParent @@ -1135,6 +1136,10 @@ export function MessageTimeline(props: { const m = messageByID().get(userMessageRow().userMessageID) if (m?.role === "user") return m }) + const messageComments = createMemo(() => { + if (!settings.general.newLayoutDesigns()) return [] + return getMsgParts(userMessageRow().userMessageID).flatMap((part) => MessageComment.fromPart(part) ?? []) + }) return ( @@ -1146,6 +1151,7 @@ export function MessageTimeline(props: { parts={getMsgParts(userMessageRow().userMessageID)} actions={props.actions} useV2Actions={settings.general.newLayoutDesigns()} + comments={messageComments()} />
@@ -1243,12 +1249,13 @@ export function MessageTimeline(props: { const initialRow = timelineRowByKey().get(props.rowKey)! const item = createMemo(() => virtualItemByKey().get(props.rowKey) ?? initialItem) const row = createMemo(() => timelineRowByKey().get(props.rowKey) ?? initialRow) - const asyncFile = () => { + const tool = () => { const value = row() - if (value._tag !== "AssistantPart" || value.group.type !== "part") return false + if (value._tag !== "AssistantPart" || value.group.type !== "part") return const part = getMsgPart(value.group.ref.messageID, value.group.ref.partID) - return part?.type === "tool" && ["edit", "write", "apply_patch"].includes(part.tool) + if (part?.type === "tool") return part } + const asyncFile = () => ["edit", "write", "apply_patch"].includes(tool()?.tool ?? "") const [ready, setReady] = createSignal(initialItem.size <= timelineFallbackItemSize || !asyncFile()) let contentMeasureFrame: number | undefined @@ -1278,6 +1285,8 @@ export function MessageTimeline(props: { width: "100%", height: `${item().size}px`, overflow: "clip", + // Rounded virtual measurements can otherwise clip a framed row's outer paint. + "overflow-clip-margin": row()._tag === "TurnGap" ? undefined : "0.5px", }} >
Part[] status: Accessor showReasoningSummaries: Accessor + inlineComments: Accessor }) { const messageByID = createMemo(() => new Map(input.messages().map((message) => [message.id, message] as const))) const assistantMessagesByParent = createMemo(() => { @@ -59,6 +60,7 @@ export function createTimelineProjection(input: { input.showReasoningSummaries(), input.status().type, activeMessageID() === userMessage.id, + input.inlineComments(), ), ), ), diff --git a/packages/app/src/pages/session/timeline/rows.ts b/packages/app/src/pages/session/timeline/rows.ts index 6a9cb669d07d..72cd9a180c00 100644 --- a/packages/app/src/pages/session/timeline/rows.ts +++ b/packages/app/src/pages/session/timeline/rows.ts @@ -38,6 +38,8 @@ export namespace Timeline { showReasoning: boolean, status: SessionStatus["type"], isActive: boolean, + // v2 renders comments inside the user message attachments row instead of a strip row + inlineComments: boolean, ) { const rows: TimelineRow.TimelineRow[] = [] @@ -74,7 +76,7 @@ export namespace Timeline { : groupParts(assistantPartRefs).map((group) => ({ type: "part" as const, group })) if (previousUserMessage) rows.push(new TimelineRow.TurnGap({ userMessageID: userMessage.id })) - if (comments.length > 0) + if (comments.length > 0 && !inlineComments) rows.push( new TimelineRow.CommentStrip({ userMessageID: userMessage.id, @@ -84,7 +86,7 @@ export namespace Timeline { rows.push( new TimelineRow.UserMessage({ userMessageID: userMessage.id, - anchor: comments.length === 0, + anchor: inlineComments || comments.length === 0, }), ) diff --git a/packages/app/src/pages/session/use-composer-commands.tsx b/packages/app/src/pages/session/use-composer-commands.tsx index 98239010f5c9..9b75c7bf4cb5 100644 --- a/packages/app/src/pages/session/use-composer-commands.tsx +++ b/packages/app/src/pages/session/use-composer-commands.tsx @@ -1,6 +1,6 @@ import { useCommand, type CommandOption } from "@/context/command" import { useLanguage } from "@/context/language" -import { useLocal } from "@/context/local" +import { useLocal, type ModelSelection } from "@/context/local" import { useSettings } from "@/context/settings" import { useDialog } from "@opencode-ai/ui/context/dialog" import { getCursorPosition, setCursorPosition } from "@/components/prompt-input/editor-dom" @@ -14,7 +14,7 @@ const withCategory = (category: string) => { }) } -export const useComposerCommands = () => { +export const useComposerCommands = (input: { model?: ModelSelection } = {}) => { const command = useCommand() const dialog = useDialog() const language = useLanguage() @@ -22,6 +22,7 @@ export const useComposerCommands = () => { const settings = useSettings() const { sessionKey } = useSessionLayout() const sessionOwnership = createSessionOwnership(sessionKey) + const model = input.model ?? local.model const modelCommand = withCategory(language.t("command.category.model")) const agentCommand = withCategory(language.t("command.category.agent")) @@ -43,7 +44,7 @@ export const useComposerCommands = () => { } const { DialogSelectModel } = await import("@/components/dialog-select-model") owner.run(() => { - void dialog.show(() => , restoreComposer) + void dialog.show(() => , restoreComposer) }) } @@ -61,7 +62,7 @@ export const useComposerCommands = () => { title: language.t("command.model.variant.cycle"), description: language.t("command.model.variant.cycle.description"), keybind: "shift+mod+d", - onSelect: () => local.model.variant.cycle(), + onSelect: () => model.variant.cycle(), }), agentCommand({ id: "agent.cycle", diff --git a/packages/app/src/pages/session/v2/review-panel-v2.tsx b/packages/app/src/pages/session/v2/review-panel-v2.tsx index a202a1e39dae..143b4f5f528a 100644 --- a/packages/app/src/pages/session/v2/review-panel-v2.tsx +++ b/packages/app/src/pages/session/v2/review-panel-v2.tsx @@ -5,7 +5,6 @@ import { SESSION_REVIEW_V2_SIDEBAR_WIDTH_MIN, SessionReviewV2, SessionReviewV2Sidebar, - SessionReviewV2SidebarToggle, } from "@opencode-ai/session-ui/v2/session-review-v2" import { SessionReviewFilePreviewV2 } from "@opencode-ai/session-ui/v2/session-review-file-preview-v2" import { DiffChanges } from "@opencode-ai/ui/v2/diff-changes-v2" @@ -66,6 +65,8 @@ export function ReviewPanelV2(props: ReviewPanelV2Props) { ) const searching = createMemo(() => props.state.filter().trim().length > 0) const kinds = createMemo(() => reviewDiffKinds(diffs())) + // Changes-only trees omit "M" — every row is already a change; A/D stay visible. + const treeKinds = createMemo(() => new Map([...kinds()].filter(([, kind]) => kind !== "mix"))) const activeDiff = createMemo(() => { // A focused comment takes over the preview until the preview applies it and // clears the focus; the owner then persists the file as the active selection. @@ -113,9 +114,6 @@ export function ReviewPanelV2(props: ReviewPanelV2Props) { stats={} empty={props.empty} sidebarOpen={props.state.sidebarOpened()} - sidebarToggle={ - - } sidebar={ // Always mounted: the sidebar header hosts the changes-mode dropdown, // which must stay reachable when the current mode has zero diffs. @@ -127,7 +125,7 @@ export function ReviewPanelV2(props: ReviewPanelV2Props) { diffs={diffs} filteredFiles={filteredFiles} searching={searching} - kinds={kinds} + kinds={treeKinds} activeDiff={activeDiff} /> } diff --git a/packages/app/src/pages/session/v2/session-file-browser-tab.tsx b/packages/app/src/pages/session/v2/session-file-browser-tab.tsx index 597ebd3f7edc..04ad1e3c8253 100644 --- a/packages/app/src/pages/session/v2/session-file-browser-tab.tsx +++ b/packages/app/src/pages/session/v2/session-file-browser-tab.tsx @@ -1,14 +1,9 @@ import { createMemo, createSignal, createUniqueId, Show } from "solid-js" import { createQuery } from "@tanstack/solid-query" -import { Tabs } from "@opencode-ai/ui/tabs" import { Icon } from "@opencode-ai/ui/icon" -import { - SessionFilePanelV2, - SessionFilePanelV2Empty, - SessionFilePanelV2Title, -} from "@opencode-ai/session-ui/v2/session-file-panel-v2" -import { SessionReviewV2Sidebar, SessionReviewV2SidebarToggle } from "@opencode-ai/session-ui/v2/session-review-v2" -import FileTree, { type Kind } from "@/components/file-tree" +import { SessionFilePanelV2, SessionFilePanelV2Empty } from "@opencode-ai/session-ui/v2/session-file-panel-v2" +import { SessionReviewV2Sidebar } from "@opencode-ai/session-ui/v2/session-review-v2" +import FileTreeV2, { type Kind } from "@/components/file-tree-v2" import { useFile } from "@/context/file" import { useLanguage } from "@/context/language" import { useLayout } from "@/context/layout" @@ -46,6 +41,7 @@ export function SessionFileBrowserTab(props: { const resultsID = `session-file-browser-results-${createUniqueId()}` const [filter, setFilter] = createSignal("") const [explicitHighlight, setExplicitHighlight] = createSignal() + const sidebarOpened = () => props.placeholder || props.state.sidebarOpened() const query = createMemo(() => filter().trim()) const search = createQuery(() => { const value = query() @@ -92,102 +88,92 @@ export function SessionFileBrowserTab(props: { }) } + // Keep the sidebar outside Kobalte Tabs.Content: a morphing content value + // unmounts the whole panel on every file-tab switch and resets sidebar scroll. return ( - - - - - {title()} - - - } - sidebar={ - {title()}} - filter={filter()} - onFilterChange={setFilter} - onFilterKeyDown={onFilterKeyDown} - filterAutofocus={props.placeholder} - filterRef={props.filterRef} - filterControls={resultsID} - filterActiveDescendant={highlighted() ? optionID(highlighted()!) : undefined} - filterExpanded={query().length > 0 && files().length > 0} - width={props.state.sidebarWidth()} - onWidthChange={props.state.resizeSidebar} + {title()}} + filter={filter()} + onFilterChange={setFilter} + onFilterKeyDown={onFilterKeyDown} + filterAutofocus={props.placeholder} + filterRef={props.filterRef} + filterControls={resultsID} + filterActiveDescendant={highlighted() ? optionID(highlighted()!) : undefined} + filterExpanded={query().length > 0 && files().length > 0} + width={props.state.sidebarWidth()} + onWidthChange={props.state.resizeSidebar} + > + props.onSelect(node.path)} + onFileDoubleClick={(node) => props.onSelectPermanent(node.path)} + /> + } > props.onSelect(node.path)} - onFileDoubleClick={(node) => props.onSelectPermanent(node.path)} - /> +
+ {language.t("common.loading")} + {language.t("common.loading.ellipsis")} +
} > 0} fallback={
- {language.t("common.loading")} - {language.t("common.loading.ellipsis")} + {language.t("palette.empty")}
} > - 0} - fallback={ -
- {language.t("palette.empty")} -
- } - > - { - setExplicitHighlight(path) - props.onSelect(path) - }} - onFileDoubleClick={props.onSelectPermanent} - /> -
+ { + setExplicitHighlight(path) + props.onSelect(path) + }} + onFileDoubleClick={props.onSelectPermanent} + />
-
+ + + } + > + +
+ +
{language.t("command.file.open")}
+
{language.t("session.files.selectToOpen")}
+
+ } > - -
- -
{language.t("command.file.open")}
-
{language.t("session.files.selectToOpen")}
-
- - } - > -
- - {(tab) => } - -
-
-
-
+
+ + {(tab) => } + +
+ + ) } diff --git a/packages/app/src/utils/persist.ts b/packages/app/src/utils/persist.ts index 671b798f6118..a2daae4866bc 100644 --- a/packages/app/src/utils/persist.ts +++ b/packages/app/src/utils/persist.ts @@ -463,7 +463,7 @@ function localStorageDirect(): SyncStorage { } } -const DRAFT_PERSISTED_KEYS = ["prompt", "comments", "model-selection", "file-view", "layout"] +const DRAFT_PERSISTED_KEYS = ["prompt", "comments", "file-view", "layout"] export function draftPersistedKeys() { return DRAFT_PERSISTED_KEYS diff --git a/packages/cli/package.json b/packages/cli/package.json index dd660aa2eec9..e9f1fc74fd92 100644 --- a/packages/cli/package.json +++ b/packages/cli/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/cli", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "bin": { diff --git a/packages/codemode/package.json b/packages/codemode/package.json index 263daa6a5331..a7e2ce3f4b7c 100644 --- a/packages/codemode/package.json +++ b/packages/codemode/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/codemode", - "version": "1.17.18", + "version": "1.17.19", "description": "Effect-native confined code execution over schema-described tools", "private": true, "type": "module", diff --git a/packages/codemode/sst-env.d.ts b/packages/codemode/sst-env.d.ts new file mode 100644 index 000000000000..64441936d7a0 --- /dev/null +++ b/packages/codemode/sst-env.d.ts @@ -0,0 +1,10 @@ +/* This file is auto-generated by SST. Do not edit. */ +/* tslint:disable */ +/* eslint-disable */ +/* deno-fmt-ignore-file */ +/* biome-ignore-all lint: auto-generated */ + +/// + +import "sst" +export {} \ No newline at end of file diff --git a/packages/console/app/package.json b/packages/console/app/package.json index 35d31f934297..3db6b14fc354 100644 --- a/packages/console/app/package.json +++ b/packages/console/app/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/console-app", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "scripts": { diff --git a/packages/console/app/public/robots.txt b/packages/console/app/public/robots.txt index bddac69deae1..12850523e5ee 100644 --- a/packages/console/app/public/robots.txt +++ b/packages/console/app/public/robots.txt @@ -3,4 +3,7 @@ Allow: / # Disallow shared content pages Disallow: /s/ -Disallow: /share/ \ No newline at end of file +Disallow: /share/ + +Sitemap: https://opencode.ai/sitemap.xml +Sitemap: https://opencode.ai/data/sitemap.xml diff --git a/packages/console/app/src/routes/zen/util/provider/openai.ts b/packages/console/app/src/routes/zen/util/provider/openai.ts index 0e9992a28976..cebc78a127a3 100644 --- a/packages/console/app/src/routes/zen/util/provider/openai.ts +++ b/packages/console/app/src/routes/zen/util/provider/openai.ts @@ -4,6 +4,7 @@ type Usage = { input_tokens?: number input_tokens_details?: { cached_tokens?: number + cache_write_tokens?: number } output_tokens?: number output_tokens_details?: { @@ -48,12 +49,13 @@ export const openaiHelper: ProviderHelper = ({ workspaceID }) => ({ const outputTokens = usage.output_tokens ?? 0 const reasoningTokens = usage.output_tokens_details?.reasoning_tokens ?? undefined const cacheReadTokens = usage.input_tokens_details?.cached_tokens ?? undefined + const cacheWriteTokens = usage.input_tokens_details?.cache_write_tokens ?? undefined return { inputTokens: inputTokens - (cacheReadTokens ?? 0), outputTokens, reasoningTokens, cacheReadTokens, - cacheWrite5mTokens: undefined, + cacheWrite5mTokens: cacheWriteTokens, cacheWrite1hTokens: undefined, } }, diff --git a/packages/console/app/test/providerUsage.test.ts b/packages/console/app/test/providerUsage.test.ts index d39c9fa8617b..7be3aacffddb 100644 --- a/packages/console/app/test/providerUsage.test.ts +++ b/packages/console/app/test/providerUsage.test.ts @@ -65,4 +65,20 @@ describe("provider usage extraction", () => { output_tokens: 7, }) }) + + test("parses OpenAI stream cache write usage", () => { + const usageParser = providers.openai.createUsageParser() + usageParser.parse( + 'event: response.completed\ndata: {"response":{"usage":{"input_tokens":10,"input_tokens_details":{"cached_tokens":4,"cache_write_tokens":3},"output_tokens":2}}}', + ) + + expect(providers.openai.normalizeUsage(usageParser.retrieve())).toEqual({ + inputTokens: 6, + outputTokens: 2, + reasoningTokens: undefined, + cacheReadTokens: 4, + cacheWrite5mTokens: 3, + cacheWrite1hTokens: undefined, + }) + }) }) diff --git a/packages/console/core/package.json b/packages/console/core/package.json index ffc7a756d241..66be843c0590 100644 --- a/packages/console/core/package.json +++ b/packages/console/core/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/console-core", - "version": "1.17.18", + "version": "1.17.19", "private": true, "type": "module", "license": "MIT", diff --git a/packages/console/function/package.json b/packages/console/function/package.json index 47ae1efd7a82..08c0a5604900 100644 --- a/packages/console/function/package.json +++ b/packages/console/function/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/console-function", - "version": "1.17.18", + "version": "1.17.19", "$schema": "https://json.schemastore.org/package.json", "private": true, "type": "module", diff --git a/packages/console/mail/package.json b/packages/console/mail/package.json index d0f16d771f83..1ca0ed2b7181 100644 --- a/packages/console/mail/package.json +++ b/packages/console/mail/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/console-mail", - "version": "1.17.18", + "version": "1.17.19", "dependencies": { "@jsx-email/all": "2.2.3", "@jsx-email/cli": "1.4.3", diff --git a/packages/console/support/package.json b/packages/console/support/package.json index 159a59722b50..c9d0cc9b8a65 100644 --- a/packages/console/support/package.json +++ b/packages/console/support/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/console-support", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "scripts": { diff --git a/packages/core/package.json b/packages/core/package.json index 55c74f9a24a4..4699b3e49136 100644 --- a/packages/core/package.json +++ b/packages/core/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "@opencode-ai/core", "type": "module", "license": "MIT", @@ -75,7 +75,7 @@ "@ai-sdk/google-vertex": "4.0.128", "@ai-sdk/groq": "3.0.31", "@ai-sdk/mistral": "3.0.27", - "@ai-sdk/openai": "3.0.53", + "@ai-sdk/openai": "3.0.84", "@ai-sdk/openai-compatible": "2.0.41", "@ai-sdk/perplexity": "3.0.26", "@ai-sdk/provider": "3.0.8", @@ -110,7 +110,7 @@ "drizzle-orm": "catalog:", "effect": "catalog:", "fuzzysort": "3.1.0", - "gitlab-ai-provider": "6.10.0", + "gitlab-ai-provider": "6.11.0", "glob": "13.0.5", "google-auth-library": "10.5.0", "gray-matter": "4.0.3", diff --git a/packages/core/src/filesystem/search.ts b/packages/core/src/filesystem/search.ts index 72c9128cfd16..c7738388bcc0 100644 --- a/packages/core/src/filesystem/search.ts +++ b/packages/core/src/filesystem/search.ts @@ -128,6 +128,8 @@ export const fffLayer = Layer.effect( Fff.create({ basePath: location.directory, aiMode: true, + disableMmapCache: true, + disableContentIndexing: true, }), catch: (cause) => cause, }).pipe( diff --git a/packages/core/src/models-dev.ts b/packages/core/src/models-dev.ts index c2325cb57fda..602133d9047a 100644 --- a/packages/core/src/models-dev.ts +++ b/packages/core/src/models-dev.ts @@ -44,6 +44,21 @@ const Cost = Schema.Struct({ ), }) +const ReasoningOption = Schema.Union([ + Schema.Struct({ + type: Schema.Literal("effort"), + values: Schema.Array(Schema.NullOr(Schema.String)), + }), + Schema.Struct({ + type: Schema.Literal("toggle"), + }), + Schema.Struct({ + type: Schema.Literal("budget_tokens"), + min: Schema.optional(Schema.Finite), + max: Schema.optional(Schema.Finite), + }), +]) + export const Model = Schema.Struct({ id: Schema.String, name: Schema.String, @@ -53,6 +68,7 @@ export const Model = Schema.Struct({ reasoning: Schema.Boolean, temperature: Schema.Boolean, tool_call: Schema.Boolean, + reasoning_options: Schema.optional(Schema.Array(ReasoningOption)), interleaved: Schema.optional( Schema.Union([ Schema.Literal(true), diff --git a/packages/desktop/package.json b/packages/desktop/package.json index ece963cbc215..ab0be7d0d8f8 100644 --- a/packages/desktop/package.json +++ b/packages/desktop/package.json @@ -1,7 +1,7 @@ { "name": "@opencode-ai/desktop", "private": true, - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "homepage": "https://opencode.ai", diff --git a/packages/desktop/src/main/ipc.ts b/packages/desktop/src/main/ipc.ts index 073d288e5cf0..d6bc5bb3a83a 100644 --- a/packages/desktop/src/main/ipc.ts +++ b/packages/desktop/src/main/ipc.ts @@ -184,6 +184,16 @@ export function registerIpcHandlers(deps: Deps) { }) }) + ipcMain.handle("reveal-path", async (_event: IpcMainInvokeEvent, path: string) => { + const exists = await stat(path).then( + () => true, + () => false, + ) + if (!exists) return false + shell.showItemInFolder(path) + return true + }) + ipcMain.handle("read-clipboard-image", () => { const image = clipboard.readImage() if (image.isEmpty()) return null diff --git a/packages/desktop/src/preload/index.ts b/packages/desktop/src/preload/index.ts index 47a757a557d3..a12864c559b6 100644 --- a/packages/desktop/src/preload/index.ts +++ b/packages/desktop/src/preload/index.ts @@ -95,6 +95,7 @@ const api: ElectronAPI = { saveFilePicker: (opts) => ipcRenderer.invoke("save-file-picker", opts), openLink: (url) => ipcRenderer.send("open-link", url), openPath: (path, app) => ipcRenderer.invoke("open-path", path, app), + revealPath: (path) => ipcRenderer.invoke("reveal-path", path), readClipboardImage: () => ipcRenderer.invoke("read-clipboard-image"), showNotification: (title, body) => ipcRenderer.send("show-notification", title, body), getWindowFocused: () => ipcRenderer.invoke("get-window-focused"), diff --git a/packages/desktop/src/preload/types.ts b/packages/desktop/src/preload/types.ts index b57ac83e7f9e..b94acd661541 100644 --- a/packages/desktop/src/preload/types.ts +++ b/packages/desktop/src/preload/types.ts @@ -86,6 +86,7 @@ export type ElectronAPI = { saveFilePicker: (opts?: { title?: string; defaultPath?: string }) => Promise openLink: (url: string) => void openPath: (path: string, app?: string) => Promise + revealPath: (path: string) => Promise readClipboardImage: () => Promise<{ buffer: ArrayBuffer; width: number; height: number } | null> showNotification: (title: string, body?: string) => void getWindowFocused: () => Promise diff --git a/packages/desktop/src/renderer/index.tsx b/packages/desktop/src/renderer/index.tsx index 966fb0c0bf3a..2f14f4348144 100644 --- a/packages/desktop/src/renderer/index.tsx +++ b/packages/desktop/src/renderer/index.tsx @@ -218,6 +218,9 @@ const createPlatform = (windowState: DesktopWindowState): Platform => { } return window.api.openPath(path, app) }, + async revealPath(path: string) { + return window.api.revealPath(path) + }, back() { window.history.back() diff --git a/packages/effect-drizzle-sqlite/package.json b/packages/effect-drizzle-sqlite/package.json index 6aebc6f41d57..1a9a7c13e641 100644 --- a/packages/effect-drizzle-sqlite/package.json +++ b/packages/effect-drizzle-sqlite/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "@opencode-ai/effect-drizzle-sqlite", "type": "module", "license": "MIT", diff --git a/packages/effect-sqlite-node/package.json b/packages/effect-sqlite-node/package.json index 117aab177fed..aa7367391d6a 100644 --- a/packages/effect-sqlite-node/package.json +++ b/packages/effect-sqlite-node/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "@opencode-ai/effect-sqlite-node", "type": "module", "license": "MIT", diff --git a/packages/enterprise/package.json b/packages/enterprise/package.json index ea959c4e4f04..7d3f07b88d5f 100644 --- a/packages/enterprise/package.json +++ b/packages/enterprise/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/enterprise", - "version": "1.17.18", + "version": "1.17.19", "private": true, "type": "module", "license": "MIT", diff --git a/packages/function/package.json b/packages/function/package.json index 282076fa652e..eee06e20eb45 100644 --- a/packages/function/package.json +++ b/packages/function/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/function", - "version": "1.17.18", + "version": "1.17.19", "$schema": "https://json.schemastore.org/package.json", "private": true, "type": "module", diff --git a/packages/http-recorder/package.json b/packages/http-recorder/package.json index 534ec4bc0ff7..68a77cc19ae8 100644 --- a/packages/http-recorder/package.json +++ b/packages/http-recorder/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "@opencode-ai/http-recorder", "description": "Record and replay Effect HTTP client traffic with deterministic cassettes", "type": "module", diff --git a/packages/llm/package.json b/packages/llm/package.json index 91103fd4d090..01ad3e628dbd 100644 --- a/packages/llm/package.json +++ b/packages/llm/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "@opencode-ai/llm", "type": "module", "license": "MIT", diff --git a/packages/opencode/package.json b/packages/opencode/package.json index 89574054c37a..b0cd278f8bde 100644 --- a/packages/opencode/package.json +++ b/packages/opencode/package.json @@ -1,6 +1,6 @@ { "$schema": "https://json.schemastore.org/package.json", - "version": "1.17.18", + "version": "1.17.19", "name": "opencode", "type": "module", "license": "MIT", @@ -67,7 +67,7 @@ "@ai-sdk/google-vertex": "4.0.128", "@ai-sdk/groq": "3.0.31", "@ai-sdk/mistral": "3.0.27", - "@ai-sdk/openai": "3.0.53", + "@ai-sdk/openai": "3.0.84", "@ai-sdk/openai-compatible": "2.0.41", "@ai-sdk/perplexity": "3.0.26", "@ai-sdk/provider": "3.0.8", @@ -120,7 +120,7 @@ "drizzle-orm": "catalog:", "effect": "catalog:", "fuzzysort": "3.1.0", - "gitlab-ai-provider": "6.10.0", + "gitlab-ai-provider": "6.11.0", "glob": "13.0.5", "google-auth-library": "10.5.0", "gray-matter": "4.0.3", diff --git a/packages/opencode/src/account/account.ts b/packages/opencode/src/account/account.ts index af8ef761eb2c..4b49d2a74890 100644 --- a/packages/opencode/src/account/account.ts +++ b/packages/opencode/src/account/account.ts @@ -348,6 +348,18 @@ const layer: Layer.Layer + group.orgs.map((org) => ({ accountID: group.account.id, orgID: org.id })), + )[0] + if (!next) return + yield* repo.use(next.accountID, Option.some(next.orgID)) + }) + const config = Effect.fn("Account.config")(function* (accountID: AccountID, orgID: OrgID) { const resolved = yield* resolveAccess(accountID) if (Option.isNone(resolved)) return Option.none() @@ -445,7 +457,7 @@ const layer: Layer.Layer entry[1] !== undefined), + ), + }) const client = Rpc.client(worker) const reload = () => { client.call("reload", undefined).catch(() => {}) diff --git a/packages/opencode/src/plugin/openai/codex.ts b/packages/opencode/src/plugin/openai/codex.ts index f14d1269769e..10186635be49 100644 --- a/packages/opencode/src/plugin/openai/codex.ts +++ b/packages/opencode/src/plugin/openai/codex.ts @@ -6,12 +6,15 @@ import { setTimeout as sleep } from "node:timers/promises" import { createServer } from "http" import { OpenAIWebSocketPool } from "./ws-pool" import { OauthCallbackPage } from "@opencode-ai/core/oauth/page" +import { isRecord } from "@/util/record" const CLIENT_ID = "app_EMoamEEZ73f0CkXaXp7hrann" const ISSUER = "https://auth.openai.com" const CODEX_API_ENDPOINT = "https://chatgpt.com/backend-api/codex/responses" const OAUTH_PORT = 1455 const OAUTH_POLLING_SAFETY_MARGIN_MS = 3000 +const CODEX_COMPATIBILITY_VERSION = "0.144.0" +const RESPONSES_LITE_MODEL = "gpt-5.6-luna" const ALLOWED_MODELS = new Set(["gpt-5.5", "gpt-5.3-codex-spark", "gpt-5.4", "gpt-5.4-mini"]) const DISALLOWED_MODELS = new Set(["gpt-5.5-pro"]) @@ -283,8 +286,10 @@ export async function CodexAuthPlugin(input: PluginInput, options: CodexAuthPlug return Object.fromEntries( Object.entries(provider.models) .filter(([, model]) => { + if (model.options.reasoningMode === "pro") return false if (ALLOWED_MODELS.has(model.api.id)) return true if (DISALLOWED_MODELS.has(model.api.id)) return false + if (model.api.id === "gpt-5.6") return false const match = model.api.id.match(/^gpt-(\d+\.\d+)/) return match ? parseFloat(match[1]) > 5.4 : false }) @@ -303,7 +308,13 @@ export async function CodexAuthPlugin(input: PluginInput, options: CodexAuthPlug input: 272_000, output: 128_000, } - : model.limit, + : model.id.includes("gpt-5.6") + ? { + context: 500_000, + input: 372_000, + output: 128_000, + } + : model.limit, }, ]), ) @@ -410,6 +421,9 @@ export async function CodexAuthPlugin(input: PluginInput, options: CodexAuthPlug const requestInit = { ...init, + body: parsed.pathname.endsWith("/responses") + ? prepareResponsesLiteRequest(init?.body, headers) + : init?.body, headers, } if (websocketFetch && parsed.pathname.endsWith("/responses")) return websocketFetch(url, requestInit) @@ -554,3 +568,61 @@ export async function CodexAuthPlugin(input: PluginInput, options: CodexAuthPlug }, } } + +function prepareResponsesLiteRequest(body: BodyInit | null | undefined, headers: Headers) { + if (typeof body !== "string") return body + const request: unknown = JSON.parse(body) + if (!isRecord(request)) return body + if (request.model !== RESPONSES_LITE_MODEL) return body + if (!Array.isArray(request.input)) throw new Error("Responses Lite requires an input array") + if (request.tools !== undefined && !Array.isArray(request.tools)) { + throw new Error("Responses Lite requires a tools array") + } + if (request.instructions !== undefined && typeof request.instructions !== "string") { + throw new Error("Responses Lite requires string instructions") + } + + const sessionID = headers.get("session-id") + if (!sessionID) throw new Error("Responses Lite requires a session-id header") + + function stripImageDetail(value: unknown) { + if (Array.isArray(value)) { + value.forEach(stripImageDetail) + return + } + if (!isRecord(value)) return + if (value.type === "input_image") delete value.detail + Object.values(value).forEach(stripImageDetail) + } + + stripImageDetail(request.input) + request.input = [ + { type: "additional_tools", role: "developer", tools: request.tools ?? [] }, + ...(request.instructions + ? [ + { + type: "message", + role: "developer", + content: [{ type: "input_text", text: request.instructions }], + }, + ] + : []), + ...request.input, + ] + delete request.tools + delete request.instructions + request.tool_choice = "auto" + request.parallel_tool_calls = false + request.prompt_cache_key = sessionID + request.reasoning = { + ...(isRecord(request.reasoning) ? request.reasoning : {}), + context: "all_turns", + } + + headers.set("session-id", sessionID) + headers.set("x-session-affinity", sessionID) + headers.set("version", CODEX_COMPATIBILITY_VERSION) + headers.set(OpenAIWebSocketPool.RESPONSES_LITE_HEADER, "true") + headers.delete("content-length") + return JSON.stringify(request) +} diff --git a/packages/opencode/src/plugin/openai/ws-pool.ts b/packages/opencode/src/plugin/openai/ws-pool.ts index 3cbb29a3012a..2cc7dcc94011 100644 --- a/packages/opencode/src/plugin/openai/ws-pool.ts +++ b/packages/opencode/src/plugin/openai/ws-pool.ts @@ -4,6 +4,8 @@ import { isRecord } from "@/util/record" import { OpenAIWebSocket } from "./ws" export const TITLE_HEADER = "x-opencode-title" +export const RESPONSES_LITE_HEADER = "x-openai-internal-codex-responses-lite" +const RESPONSES_LITE_CLIENT_METADATA = "ws_request_header_x_openai_internal_codex_responses_lite" export interface CreateWebSocketFetchOptions { httpFetch?: typeof globalThis.fetch @@ -98,7 +100,16 @@ export function createWebSocketFetch(options?: CreateWebSocketFetchOptions) { }) const response = OpenAIWebSocket.streamResponsesWebSocket({ socket: entry.socket, - body, + body: + internalHeaders[RESPONSES_LITE_HEADER] === "true" + ? { + ...body, + client_metadata: { + ...(isRecord(body.client_metadata) ? body.client_metadata : {}), + [RESPONSES_LITE_CLIENT_METADATA]: "true", + }, + } + : body, idleTimeout, signal: init?.signal ?? undefined, onFirstEvent: (error) => resolveFirstEvent(error ?? true), diff --git a/packages/opencode/src/provider/provider.ts b/packages/opencode/src/provider/provider.ts index c0a222d649bb..2b1b53c231bb 100644 --- a/packages/opencode/src/provider/provider.ts +++ b/packages/opencode/src/provider/provider.ts @@ -1247,9 +1247,11 @@ function fromModelsDevModel(provider: ModelsDev.Provider, model: ModelsDev.Model variants: {}, } + const variants = ProviderTransform.reasoningVariants(model, base) ?? ProviderTransform.variants(base) + return { ...base, - variants: mapValues(ProviderTransform.variants(base), (v) => v), + variants: mapValues(variants, (v) => v), } } @@ -1265,14 +1267,7 @@ export function fromModelsDevProvider(provider: ModelsDev.Provider): Info { id: ModelV2.ID.make(id), name: `${model.name} ${mode[0].toUpperCase()}${mode.slice(1)}`, cost: opts.cost ? mergeDeep(base.cost, cost(opts.cost)) : base.cost, - options: opts.provider?.body - ? Object.fromEntries( - Object.entries(opts.provider.body).map(([k, v]) => [ - k.replace(/_([a-z])/g, (_, c) => c.toUpperCase()), - v, - ]), - ) - : base.options, + options: modeOptions(base, opts.provider?.body), headers: opts.provider?.headers ?? base.headers, } } @@ -1287,6 +1282,17 @@ export function fromModelsDevProvider(provider: ModelsDev.Provider): Info { } } +function modeOptions(model: Model, body: Record | undefined) { + if (!body) return model.options + const options = Object.fromEntries( + Object.entries(body).map(([key, value]) => [key.replace(/_([a-z])/g, (_, char) => char.toUpperCase()), value]), + ) + const reasoning = body.reasoning + if (model.api.npm !== "@ai-sdk/openai" || !isRecord(reasoning) || typeof reasoning.mode !== "string") return options + const { reasoning: _, ...rest } = options + return { ...rest, reasoningMode: reasoning.mode } +} + function modelSuggestions(provider: Info | undefined, modelID: ModelV2.ID, enableExperimentalModels: boolean) { const available = provider ? Object.keys(provider.models).filter((id) => { @@ -1492,7 +1498,11 @@ const layer = Layer.effect( release_date: model.release_date ?? existingModel?.release_date ?? "", variants: {}, } - const merged = mergeDeep(ProviderTransform.variants(parsedModel), model.variants ?? {}) + const variants = + existingModel?.api.npm === parsedModel.api.npm + ? (existingModel.variants ?? ProviderTransform.variants(parsedModel)) + : ProviderTransform.variants(parsedModel) + const merged = mergeDeep(variants, model.variants ?? {}) parsedModel.variants = mapValues( pickBy(merged, (v) => !v.disabled), (v) => omit(v, ["disabled"]), @@ -1620,7 +1630,7 @@ const layer = Layer.effect( ) delete provider.models[modelID] - if (!model.variants || Object.keys(model.variants).length === 0) { + if (model.variants === undefined) { model.variants = mapValues(ProviderTransform.variants(model), (v) => v) } diff --git a/packages/opencode/src/provider/transform.ts b/packages/opencode/src/provider/transform.ts index 20c73d021671..2b9b8d5930f4 100644 --- a/packages/opencode/src/provider/transform.ts +++ b/packages/opencode/src/provider/transform.ts @@ -759,7 +759,7 @@ export function variants(model: Provider.Model): Record [ @@ -792,8 +792,8 @@ export function variants(model: Provider.Model): Record option.type === "effort") + if (effort) return nonEmptyVariants(effortVariants(target, effort.values)) + + const toggle = options.some((option) => option.type === "toggle") + const budget = options.find((option) => option.type === "budget_tokens") + if (!budget) return toggle ? nonEmptyVariants(reasoningToggle(target)) : undefined + + return nonEmptyVariants({ + ...(toggle ? reasoningToggle(target) : {}), + ...budgetVariants(target, budget.min, budget.max), + }) +} + +function effortVariants(model: Provider.Model, values: readonly unknown[]) { + return Object.fromEntries( + values.flatMap((value) => { + const id = (() => { + if (value === null) return "none" + if (typeof value === "string") return value + })() + if (id === undefined) return [] + const settings = reasoningEffort(model, id) + return settings ? [[id, settings]] : [] + }), + ) +} + +function budgetVariants(model: Provider.Model, min?: number, max?: number) { + const maximum = Math.min(max ?? OUTPUT_TOKEN_MAX - 1, model.limit.output - 1, OUTPUT_TOKEN_MAX - 1) + if (maximum <= 0) return {} + const high = Math.min(Math.max(min ?? 0, Math.floor((maximum + 1) / 2)), maximum) + return Object.fromEntries( + [ + { id: "high", budget: high }, + { id: "max", budget: maximum }, + ].flatMap((item) => { + const settings = reasoningBudget(model, item.budget) + return settings ? [[item.id, settings]] : [] + }), + ) +} + +function nonEmptyVariants(variants: NonNullable): Provider.Model["variants"] { + return Object.keys(variants).length > 0 ? variants : undefined +} + +function reasoningToggle(model: Provider.Model): NonNullable { + if (model.api.npm === "@ai-sdk/alibaba") + return { + none: { enableThinking: false }, + high: { enableThinking: true }, + } + if (model.api.npm === "@ai-sdk/cohere") + return { + none: { thinking: { type: "disabled" } }, + high: { thinking: { type: "enabled" } }, + } + return {} +} + +function reasoningEffort(model: Provider.Model, effort: string) { + switch (model.api.npm) { + case "@openrouter/ai-sdk-provider": + return { reasoning: { effort } } + case "@ai-sdk/anthropic": + case "@ai-sdk/google-vertex/anthropic": + return anthropicEffort(model, effort) + case "@ai-sdk/google": + case "@ai-sdk/google-vertex": + return { thinkingConfig: { includeThoughts: true, thinkingLevel: effort } } + case "@ai-sdk/amazon-bedrock": + if (anthropicAdaptiveEfforts(model.api.id)) + return { + reasoningConfig: { + type: "adaptive", + maxReasoningEffort: effort, + ...(anthropicOmitsThinking(model.api.id) ? { display: "summarized" } : {}), + }, + } + if (model.api.id.includes("anthropic")) return + return { reasoningConfig: { type: "enabled", maxReasoningEffort: effort } } + case "@ai-sdk/gateway": + if (model.id.includes("anthropic")) return { thinking: { type: "adaptive", display: "summarized" }, effort } + if (model.id.includes("google")) return { thinkingConfig: { includeThoughts: true, thinkingLevel: effort } } + return { reasoningEffort: effort } + case "@ai-sdk/github-copilot": + // OAuth discovery replaces these with variants from Copilot's /models capabilities. + if (model.id.includes("gemini")) return + if (model.id.includes("claude")) return { reasoningEffort: effort } + return { reasoningEffort: effort, reasoningSummary: "auto", include: INCLUDE_ENCRYPTED_REASONING } + case "@ai-sdk/openai": + case "@ai-sdk/amazon-bedrock/mantle": + return { reasoningEffort: effort, reasoningSummary: "auto", include: INCLUDE_ENCRYPTED_REASONING } + case "@ai-sdk/azure": + return { reasoningEffort: effort, reasoningSummary: "auto", include: INCLUDE_ENCRYPTED_REASONING } + case "@jerome-benoit/sap-ai-provider-v2": + if (model.id.includes("anthropic")) + return { modelParams: { thinking: { type: "adaptive", display: "summarized" }, output_config: { effort } } } + return { modelParams: { reasoning_effort: effort } } + case "@ai-sdk/openai-compatible": + case "@ai-sdk/xai": + case "@ai-sdk/mistral": + case "@ai-sdk/groq": + case "@ai-sdk/cerebras": + case "@ai-sdk/deepinfra": + case "@ai-sdk/togetherai": + case "venice-ai-sdk-provider": + case "ai-gateway-provider": + return { reasoningEffort: effort } + case "@ai-sdk/cohere": + case "@ai-sdk/perplexity": + case "@ai-sdk/vercel": + case "@ai-sdk/alibaba": + case "gitlab-ai-provider": + return + } +} + +function anthropicEffort(model: Provider.Model, effort: string) { + if (["opus-4-5", "opus-4.5"].some((value) => model.api.id.includes(value))) return { effort } + if (!anthropicAdaptiveEfforts(model.api.id)) return + return { + thinking: { + type: "adaptive", + ...(anthropicOmitsThinking(model.api.id) ? { display: "summarized" } : {}), + }, + effort, + } +} + +function reasoningBudget(model: Provider.Model, budget: number) { + switch (model.api.npm) { + case "@openrouter/ai-sdk-provider": + return { reasoning: { max_tokens: budget } } + case "@ai-sdk/anthropic": + case "@ai-sdk/google-vertex/anthropic": + return { thinking: { type: "enabled", budgetTokens: budget } } + case "@ai-sdk/google": + case "@ai-sdk/google-vertex": + return { thinkingConfig: { includeThoughts: true, thinkingBudget: budget } } + case "@ai-sdk/amazon-bedrock": + return { reasoningConfig: { type: "enabled", budgetTokens: budget } } + case "@ai-sdk/gateway": + if (model.id.includes("anthropic")) return { thinking: { type: "enabled", budgetTokens: budget } } + if (model.id.includes("google")) return { thinkingConfig: { includeThoughts: true, thinkingBudget: budget } } + return + case "@ai-sdk/cohere": + return { thinking: { type: "enabled", tokenBudget: budget } } + case "@ai-sdk/alibaba": + return { enableThinking: true, thinkingBudget: budget } + case "@jerome-benoit/sap-ai-provider-v2": + if (model.id.includes("anthropic")) + return { modelParams: { thinking: { type: "enabled", budget_tokens: budget } } } + if (model.id.includes("gemini")) + return { modelParams: { thinkingConfig: { includeThoughts: true, thinkingBudget: budget } } } + return + case "@ai-sdk/amazon-bedrock/mantle": + case "@ai-sdk/azure": + case "@ai-sdk/cerebras": + case "@ai-sdk/deepinfra": + case "@ai-sdk/github-copilot": + case "@ai-sdk/groq": + case "@ai-sdk/mistral": + case "@ai-sdk/openai": + case "@ai-sdk/openai-compatible": + case "@ai-sdk/perplexity": + case "@ai-sdk/togetherai": + case "@ai-sdk/vercel": + case "@ai-sdk/xai": + case "ai-gateway-provider": + case "gitlab-ai-provider": + case "venice-ai-sdk-provider": + return + } +} + export * as ProviderTransform from "./transform" diff --git a/packages/opencode/test/account/service.test.ts b/packages/opencode/test/account/service.test.ts index 0ebe69c2394b..672d54971623 100644 --- a/packages/opencode/test/account/service.test.ts +++ b/packages/opencode/test/account/service.test.ts @@ -173,6 +173,53 @@ it.live("orgsByAccount groups orgs per account", () => }), ) +it.live("remove switches to another org when the active account is removed", () => + Effect.gen(function* () { + const first = AccountID.make("user-1") + const second = AccountID.make("user-2") + + yield* AccountRepo.Service.use((r) => + r.persistAccount({ + id: first, + email: "one@example.com", + url: "https://one.example.com", + accessToken: AccessToken.make("at_1"), + refreshToken: RefreshToken.make("rt_1"), + expiry: Date.now() + outsideEagerRefreshWindow, + orgID: Option.some(OrgID.make("org-1")), + }), + ) + + yield* AccountRepo.Service.use((r) => + r.persistAccount({ + id: second, + email: "two@example.com", + url: "https://two.example.com", + accessToken: AccessToken.make("at_2"), + refreshToken: RefreshToken.make("rt_2"), + expiry: Date.now() + outsideEagerRefreshWindow, + orgID: Option.some(OrgID.make("org-2")), + }), + ) + + const client = HttpClient.make((req) => + Effect.succeed( + req.url === "https://one.example.com/api/orgs" ? json(req, [org("org-1", "One")]) : json(req, [], 404), + ), + ) + + yield* Account.use.remove(second).pipe(Effect.provide(live(client))) + + const active = yield* AccountRepo.use.active() + expect(Option.getOrThrow(active)).toEqual( + expect.objectContaining({ + id: first, + active_org_id: OrgID.make("org-1"), + }), + ) + }), +) + it.live("token refresh persists the new token", () => Effect.gen(function* () { const id = AccountID.make("user-1") diff --git a/packages/opencode/test/cli/tui/thread.test.ts b/packages/opencode/test/cli/tui/thread.test.ts index 73f87d904bfd..2b2969808690 100644 --- a/packages/opencode/test/cli/tui/thread.test.ts +++ b/packages/opencode/test/cli/tui/thread.test.ts @@ -16,6 +16,12 @@ describe("tui thread", () => { expect(source).not.toContain('import("./app")') }) + test("forwards the CLI environment to the TUI worker", async () => { + const source = await Bun.file(new URL("../../../src/cli/cmd/tui.ts", import.meta.url)).text() + + expect(source).toMatch(/new Worker\(file, \{\s*env: Object\.fromEntries\(\s*Object\.entries\(process\.env\)/) + }) + async function check(project?: string) { await using tmp = await tmpdir({ git: true }) const link = path.join(path.dirname(tmp.path), path.basename(tmp.path) + "-link") diff --git a/packages/opencode/test/plugin/codex.test.ts b/packages/opencode/test/plugin/codex.test.ts index 7142bb3e20c9..1e1c190ccf78 100644 --- a/packages/opencode/test/plugin/codex.test.ts +++ b/packages/opencode/test/plugin/codex.test.ts @@ -149,6 +149,48 @@ describe("plugin.codex", () => { await enabled.dispose?.() }) + test("filters unsupported modes and uses Codex context limits for OAuth GPT models", async () => { + const hooks = await CodexAuthPlugin({} as never) + const limit = { context: 1_050_000, input: 922_000, output: 128_000 } + const provider = { + models: { + ...Object.fromEntries( + ["gpt-5.4", "gpt-5.5", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna", "gpt-5.7-pro"].map( + (id) => [id, { id, api: { id }, limit, cost: {}, options: {} }], + ), + ), + "gpt-5.4-pro": { + id: "gpt-5.4-pro", + api: { id: "gpt-5.4" }, + limit, + cost: {}, + options: { reasoningMode: "pro" }, + }, + "gpt-5.6-sol-high": { + id: "gpt-5.6-sol-high", + api: { id: "gpt-5.6-sol" }, + limit, + cost: {}, + options: { reasoningEffort: "high" }, + }, + }, + } + + const models = await hooks.provider!.models!(provider as never, { auth: { type: "oauth" } } as never) + + expect(models["gpt-5.4"]?.limit).toEqual(limit) + expect(models["gpt-5.5"]?.limit).toEqual({ context: 400_000, input: 272_000, output: 128_000 }) + expect(models["gpt-5.6-sol"]?.limit).toEqual({ context: 500_000, input: 372_000, output: 128_000 }) + expect(models["gpt-5.6-terra"]?.limit).toEqual({ context: 500_000, input: 372_000, output: 128_000 }) + expect(models["gpt-5.6-luna"]?.limit).toEqual({ context: 500_000, input: 372_000, output: 128_000 }) + expect(models["gpt-5.4-pro"]).toBeUndefined() + expect(models["gpt-5.7-pro"]).toBeDefined() + expect(models["gpt-5.6-sol-high"]).toBeDefined() + expect(await hooks.provider!.models!(provider as never, { auth: { type: "api" } } as never)).toBe( + provider.models as never, + ) + }) + test("deduplicates concurrent Codex token refreshes", async () => { let auth = { type: "oauth" as const, diff --git a/packages/opencode/test/provider/provider.test.ts b/packages/opencode/test/provider/provider.test.ts index c27877c1f7be..e656890b01b0 100644 --- a/packages/opencode/test/provider/provider.test.ts +++ b/packages/opencode/test/provider/provider.test.ts @@ -159,10 +159,10 @@ it.instance( const providers = yield* list expect(providers[ProviderV2.ID.anthropic]).toBeDefined() const models = Object.keys(providers[ProviderV2.ID.anthropic].models) - expect(models).toContain("claude-sonnet-4-20250514") + expect(models).toContain("claude-sonnet-4-6") expect(models.length).toBe(1) }), - { config: { provider: { anthropic: { whitelist: ["claude-sonnet-4-20250514"] } } } }, + { config: { provider: { anthropic: { whitelist: ["claude-sonnet-4-6"] } } } }, ) it.instance( @@ -301,10 +301,10 @@ it.instance("getModel returns model for valid provider/model", () => Effect.gen(function* () { yield* setProcessEnv("ANTHROPIC_API_KEY", "test-api-key") const provider = yield* Provider.Service - const model = yield* provider.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-20250514")) + const model = yield* provider.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-6")) expect(model).toBeDefined() expect(String(model.providerID)).toBe("anthropic") - expect(String(model.id)).toBe("claude-sonnet-4-20250514") + expect(String(model.id)).toBe("claude-sonnet-4-6") const language = yield* provider.getLanguage(model) expect(language).toBeDefined() }), @@ -435,7 +435,7 @@ it.instance( "model options are merged from existing model", Effect.gen(function* () { const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.options.customOption).toBe("custom-value") }), { @@ -443,7 +443,7 @@ it.instance( provider: { anthropic: { options: { apiKey: "test-api-key" }, - models: { "claude-sonnet-4-20250514": { options: { customOption: "custom-value" } } }, + models: { "claude-sonnet-4-6": { options: { customOption: "custom-value" } } }, }, }, }, @@ -549,7 +549,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.name).toBe("Custom Name for Sonnet") expect(model.capabilities.toolcall).toBe(true) expect(model.capabilities.attachment).toBe(true) @@ -557,7 +557,46 @@ it.instance( }), { config: { - provider: { anthropic: { models: { "claude-sonnet-4-20250514": { name: "Custom Name for Sonnet" } } } }, + provider: { anthropic: { models: { "claude-sonnet-4-6": { name: "Custom Name for Sonnet" } } } }, + }, + }, +) + +it.instance( + "model config preserves explicitly empty models.dev variants", + Effect.gen(function* () { + yield* set("OPENAI_API_KEY", "test-api-key") + const providers = yield* list + const model = providers[ProviderV2.ID.openai].models["custom-gpt-chat"] + expect(model.name).toBe("Custom GPT Chat") + expect(model.variants).toEqual({}) + }), + { + config: { + provider: { + openai: { models: { "custom-gpt-chat": { id: "gpt-5-chat-latest", name: "Custom GPT Chat" } } }, + }, + }, + }, +) + +it.instance( + "model config regenerates variants when overriding the provider package", + Effect.gen(function* () { + yield* set("ANTHROPIC_API_KEY", "test-api-key") + const providers = yield* list + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] + expect(model.variants?.low).toEqual({ reasoningEffort: "low" }) + expect(model.variants?.max).toBeUndefined() + }), + { + config: { + provider: { + anthropic: { + npm: "@ai-sdk/openai-compatible", + models: { "claude-sonnet-4-6": { name: "Claude via OpenAI" } }, + }, + }, }, }, ) @@ -590,16 +629,16 @@ it.instance( const providers = yield* list expect(providers[ProviderV2.ID.anthropic]).toBeDefined() const models = Object.keys(providers[ProviderV2.ID.anthropic].models) - expect(models).toContain("claude-sonnet-4-20250514") - expect(models).not.toContain("claude-opus-4-20250514") + expect(models).toContain("claude-sonnet-4-6") + expect(models).not.toContain("claude-opus-4-6") expect(models.length).toBe(1) }), { config: { provider: { anthropic: { - whitelist: ["claude-sonnet-4-20250514", "claude-opus-4-20250514"], - blacklist: ["claude-opus-4-20250514"], + whitelist: ["claude-sonnet-4-6", "claude-opus-4-6"], + blacklist: ["claude-opus-4-6"], }, }, }, @@ -773,9 +812,9 @@ it.instance( const model = yield* Provider.use.getSmallModel(ProviderV2.ID.anthropic) expect(model).toBeDefined() expect(String(model?.providerID)).toBe("anthropic") - expect(String(model?.id)).toBe("claude-sonnet-4-20250514") + expect(String(model?.id)).toBe("claude-sonnet-4-6") }), - { config: { small_model: "anthropic/claude-sonnet-4-20250514" } }, + { config: { small_model: "anthropic/claude-sonnet-4-6" } }, ) it.instance( @@ -1094,8 +1133,8 @@ it.instance( it.instance("getModel returns consistent results", () => Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") - const model1 = yield* Provider.use.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-20250514")) - const model2 = yield* Provider.use.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-20250514")) + const model1 = yield* Provider.use.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-6")) + const model2 = yield* Provider.use.getModel(ProviderV2.ID.anthropic, ModelV2.ID.make("claude-sonnet-4-6")) expect(model1.providerID).toEqual(model2.providerID) expect(model1.id).toEqual(model2.id) expect(model1).toEqual(model2) @@ -1337,16 +1376,17 @@ it.instance( }, ) -test("mode cost preserves over-200k pricing from base model", () => { +test("mode options and cost are derived from the base model", () => { const provider = { id: "openai", name: "OpenAI", env: [], + npm: "@ai-sdk/openai", api: "https://api.openai.com/v1", models: { - "gpt-5.4": { - id: "gpt-5.4", - name: "GPT-5.4", + "gpt-5.6-sol": { + id: "gpt-5.6-sol", + name: "GPT-5.6 Sol", family: "gpt", release_date: "2026-03-05", attachment: true, @@ -1382,18 +1422,29 @@ test("mode cost preserves over-200k pricing from base model", () => { }, }, }, + pro: { + provider: { + body: { + reasoning: { mode: "pro" }, + service_tier: "priority", + }, + }, + }, }, }, }, }, } as unknown as ModelsDev.Provider - const model = Provider.fromModelsDevProvider(provider).models["gpt-5.4-fast"] + const model = Provider.fromModelsDevProvider(provider).models["gpt-5.6-sol-fast"] expect(model.cost.input).toEqual(5) expect(model.cost.output).toEqual(30) expect(model.cost.cache.read).toEqual(0.5) expect(model.cost.cache.write).toEqual(0) expect(model.options["serviceTier"]).toEqual("priority") + const pro = Provider.fromModelsDevProvider(provider).models["gpt-5.6-sol-pro"] + expect(pro.api.id).toEqual("gpt-5.6-sol") + expect(pro.options).toEqual({ reasoningMode: "pro", serviceTier: "priority" }) expect(model.cost.experimentalOver200K).toEqual({ input: 5, output: 22.5, @@ -1426,6 +1477,62 @@ test("models.dev normalization fills required response fields", () => { expect(model.release_date).toBe("") }) +test("models.dev reasoning options replace generated variants and unsupported options fall back", () => { + const provider = { + id: "reasoning", + name: "Reasoning", + env: [], + npm: "@ai-sdk/openai", + models: { + explicit: { + id: "gpt-5.4", + name: "Explicit", + reasoning: true, + reasoning_options: [{ type: "effort", values: ["low"] }], + limit: { context: 128_000, output: 64_000 }, + }, + empty: { + id: "gpt-5.4", + name: "Empty", + reasoning: true, + reasoning_options: [], + limit: { context: 128_000, output: 64_000 }, + }, + fallback: { + id: "gpt-5.4", + name: "Fallback", + reasoning: true, + reasoning_options: [{ type: "toggle" }], + limit: { context: 128_000, output: 64_000 }, + }, + override: { + id: "gemini-3-pro", + name: "Override", + reasoning: true, + reasoning_options: [{ type: "effort", values: ["high"] }], + provider: { npm: "@ai-sdk/google" }, + limit: { context: 128_000, output: 64_000 }, + experimental: { modes: { fast: {} } }, + }, + }, + } as unknown as ModelsDev.Provider + + const models = Provider.fromModelsDevProvider(provider).models + expect(models.explicit.variants).toEqual({ + low: { + reasoningEffort: "low", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + }) + expect(models.empty.variants).toEqual({}) + expect(Object.keys(models.fallback.variants ?? {})).toEqual(["none", "low", "medium", "high", "xhigh"]) + expect(models.override.variants).toEqual({ + high: { thinkingConfig: { includeThoughts: true, thinkingLevel: "high" } }, + }) + expect(models["gemini-3-pro-fast"].variants).toEqual(models.override.variants) +}) + test("public provider info omits invalid models", () => { const provider = Provider.fromModelsDevProvider({ id: "test", @@ -1457,7 +1564,7 @@ it.instance("model variants are generated for reasoning models", () => yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list // Claude sonnet 4 has reasoning capability - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.capabilities.reasoning).toBe(true) expect(model.variants).toBeDefined() expect(Object.keys(model.variants!).length).toBeGreaterThan(0) @@ -1469,7 +1576,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.variants).toBeDefined() expect(model.variants!["high"]).toBeUndefined() // max variant should still exist @@ -1479,7 +1586,7 @@ it.instance( config: { provider: { anthropic: { - models: { "claude-sonnet-4-20250514": { variants: { high: { disabled: true } } } }, + models: { "claude-sonnet-4-6": { variants: { high: { disabled: true } } } }, }, }, }, @@ -1491,7 +1598,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.variants!["high"]).toBeDefined() expect(model.variants!["high"].thinking.budgetTokens).toBe(20000) }), @@ -1500,7 +1607,7 @@ it.instance( provider: { anthropic: { models: { - "claude-sonnet-4-20250514": { + "claude-sonnet-4-6": { variants: { high: { thinking: { type: "enabled", budgetTokens: 20000 } } }, }, }, @@ -1515,7 +1622,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.variants!["max"]).toBeDefined() expect(model.variants!["max"].disabled).toBeUndefined() expect(model.variants!["max"].customField).toBe("test") @@ -1525,7 +1632,7 @@ it.instance( provider: { anthropic: { models: { - "claude-sonnet-4-20250514": { + "claude-sonnet-4-6": { variants: { max: { disabled: false, customField: "test" } }, }, }, @@ -1540,7 +1647,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.variants).toBeDefined() expect(Object.keys(model.variants!).length).toBe(0) }), @@ -1549,8 +1656,13 @@ it.instance( provider: { anthropic: { models: { - "claude-sonnet-4-20250514": { - variants: { high: { disabled: true }, max: { disabled: true } }, + "claude-sonnet-4-6": { + variants: { + low: { disabled: true }, + medium: { disabled: true }, + high: { disabled: true }, + max: { disabled: true }, + }, }, }, }, @@ -1564,7 +1676,7 @@ it.instance( Effect.gen(function* () { yield* set("ANTHROPIC_API_KEY", "test-api-key") const providers = yield* list - const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-20250514"] + const model = providers[ProviderV2.ID.anthropic].models["claude-sonnet-4-6"] expect(model.variants!["high"]).toBeDefined() // Should have both the generated thinking config and the custom option expect(model.variants!["high"].thinking).toBeDefined() @@ -1575,7 +1687,7 @@ it.instance( provider: { anthropic: { models: { - "claude-sonnet-4-20250514": { variants: { high: { extraOption: "custom-value" } } }, + "claude-sonnet-4-6": { variants: { high: { extraOption: "custom-value" } } }, }, }, }, diff --git a/packages/opencode/test/provider/transform.test.ts b/packages/opencode/test/provider/transform.test.ts index e69dab70f65d..4c462bd3dce2 100644 --- a/packages/opencode/test/provider/transform.test.ts +++ b/packages/opencode/test/provider/transform.test.ts @@ -4,6 +4,7 @@ import { ProviderTransform } from "@/provider/transform" import { LLMRequestPrep } from "@/session/llm/request" import { ProviderV2 } from "@opencode-ai/core/provider" import { ModelV2 } from "@opencode-ai/core/model" +import { ModelsDev } from "@opencode-ai/core/models-dev" import { jsonSchema } from "ai" describe("ProviderTransform.options - setCacheKey", () => { @@ -155,6 +156,43 @@ describe("ProviderTransform.options - setCacheKey", () => { expect(result.store).toBe(false) }) + test("should set store=false for xAI provider by default", () => { + const xaiModel = { + ...mockModel, + providerID: "xai", + api: { + id: "grok-4", + url: "https://api.x.ai", + npm: "@ai-sdk/xai", + }, + } + const result = ProviderTransform.options({ + model: xaiModel, + sessionID, + providerOptions: {}, + }) + expect(result.store).toBe(false) + expect(result.promptCacheKey).toBe(sessionID) + }) + + test("should set store=false for xAI SDK regardless of provider ID", () => { + const xaiModel = { + ...mockModel, + providerID: "custom-xai", + api: { + id: "grok-4", + url: "https://api.x.ai", + npm: "@ai-sdk/xai", + }, + } + const result = ProviderTransform.options({ + model: xaiModel, + sessionID, + providerOptions: {}, + }) + expect(result.store).toBe(false) + }) + test("should set store=false for azure provider by default", () => { const azureModel = { ...mockModel, @@ -2988,6 +3026,268 @@ describe("ProviderTransform.temperature - Cohere North", () => { }) }) +describe("ProviderTransform.reasoningVariants", () => { + const model = (reasoning_options: ModelsDev.Model["reasoning_options"]) => ({ reasoning_options }) as ModelsDev.Model + const target = (npm: string, id = "test-model") => + ({ id, api: { id, npm, url: "" }, capabilities: { reasoning: true }, limit: { output: 64_000 } }) as any + + test("respects explicitly empty reasoning options", () => { + expect(ProviderTransform.reasoningVariants(model([]), target("@ai-sdk/openai"))).toEqual({}) + }) + + test.each([ + ["@openrouter/ai-sdk-provider", { reasoning: { effort: "high" } }], + ["@ai-sdk/anthropic", { thinking: { type: "adaptive" }, effort: "high" }, "claude-opus-4-6"], + [ + "@ai-sdk/google-vertex/anthropic", + { thinking: { type: "adaptive", display: "summarized" }, effort: "high" }, + "claude-opus-4-7", + ], + ["@ai-sdk/google", { thinkingConfig: { includeThoughts: true, thinkingLevel: "high" } }], + ["@ai-sdk/google-vertex", { thinkingConfig: { includeThoughts: true, thinkingLevel: "high" } }], + [ + "@ai-sdk/azure", + { + reasoningEffort: "high", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + ], + [ + "@ai-sdk/openai", + { + reasoningEffort: "high", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + ], + [ + "@ai-sdk/amazon-bedrock/mantle", + { + reasoningEffort: "high", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + ], + [ + "@ai-sdk/github-copilot", + { + reasoningEffort: "high", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + ], + ["@ai-sdk/openai-compatible", { reasoningEffort: "high" }], + ["@ai-sdk/xai", { reasoningEffort: "high" }], + ["@ai-sdk/mistral", { reasoningEffort: "high" }], + ["@ai-sdk/groq", { reasoningEffort: "high" }], + ["@ai-sdk/cerebras", { reasoningEffort: "high" }], + ["@ai-sdk/deepinfra", { reasoningEffort: "high" }], + ["@ai-sdk/togetherai", { reasoningEffort: "high" }], + ["venice-ai-sdk-provider", { reasoningEffort: "high" }], + ["ai-gateway-provider", { reasoningEffort: "high" }], + ["@ai-sdk/amazon-bedrock", { reasoningConfig: { type: "enabled", maxReasoningEffort: "high" } }], + ])("converts effort for %s", (npm, expected, ...args) => { + const id = args[0] as string | undefined + expect(ProviderTransform.reasoningVariants(model([{ type: "effort", values: ["high"] }]), target(npm, id))).toEqual( + { high: expected }, + ) + }) + + test("uses bare effort for Claude Opus 4.5", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "effort", values: ["high"] }]), + target("@ai-sdk/anthropic", "claude-opus-4-5"), + ), + ).toEqual({ high: { effort: "high" } }) + }) + + test("leaves legacy Anthropic effort options to budget fallback", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "effort", values: ["high"] }]), + target("@ai-sdk/anthropic", "claude-sonnet-4"), + ), + ).toBeUndefined() + }) + + test("uses adaptive reasoning config for Anthropic models on Bedrock", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "effort", values: ["high"] }]), + target("@ai-sdk/amazon-bedrock", "anthropic.claude-opus-4-7-v1:0"), + ), + ).toEqual({ + high: { + reasoningConfig: { + type: "adaptive", + maxReasoningEffort: "high", + display: "summarized", + }, + }, + }) + }) + + test("leaves legacy Anthropic Bedrock effort options to budget fallback", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "effort", values: ["high"] }]), + target("@ai-sdk/amazon-bedrock", "anthropic.claude-sonnet-4-v1:0"), + ), + ).toBeUndefined() + }) + + test.each([ + ["@openrouter/ai-sdk-provider", { reasoning: { max_tokens: 16_000 } }], + ["@ai-sdk/anthropic", { thinking: { type: "enabled", budgetTokens: 16_000 } }], + ["@ai-sdk/google-vertex/anthropic", { thinking: { type: "enabled", budgetTokens: 16_000 } }], + ["@ai-sdk/google", { thinkingConfig: { includeThoughts: true, thinkingBudget: 16_000 } }], + ["@ai-sdk/google-vertex", { thinkingConfig: { includeThoughts: true, thinkingBudget: 16_000 } }], + ["@ai-sdk/amazon-bedrock", { reasoningConfig: { type: "enabled", budgetTokens: 16_000 } }], + ["@ai-sdk/cohere", { thinking: { type: "enabled", tokenBudget: 16_000 } }], + ["@ai-sdk/alibaba", { enableThinking: true, thinkingBudget: 16_000 }], + ])("converts token budgets for %s", (npm, high) => { + const variants = ProviderTransform.reasoningVariants(model([{ type: "budget_tokens", min: 1_024 }]), target(npm)) + expect(variants?.high).toEqual(high) + expect(Object.keys(variants ?? {})).toEqual(["high", "max"]) + }) + + test("maps null effort to none", () => { + expect( + ProviderTransform.reasoningVariants(model([{ type: "effort", values: [null] }]), target("@ai-sdk/openai")), + ).toEqual({ + none: { + reasoningEffort: "none", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + }) + }) + + test.each([ + ["@ai-sdk/alibaba", { none: { enableThinking: false }, high: { enableThinking: true } }], + [ + "@ai-sdk/cohere", + { + none: { thinking: { type: "disabled" } }, + high: { thinking: { type: "enabled" } }, + }, + ], + ])("converts toggle options for %s", (npm, expected) => { + expect(ProviderTransform.reasoningVariants(model([{ type: "toggle" }]), target(npm))).toEqual(expected) + }) + + test("combines Cohere toggle and budget options", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "toggle" }, { type: "budget_tokens", min: 1 }]), + target("@ai-sdk/cohere"), + ), + ).toEqual({ + none: { thinking: { type: "disabled" } }, + high: { thinking: { type: "enabled", tokenBudget: 16_000 } }, + max: { thinking: { type: "enabled", tokenBudget: 31_999 } }, + }) + }) + + test("generates bounded high and max token budgets", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "budget_tokens", min: 1_024, max: 64_000 }]), + target("@ai-sdk/anthropic"), + ), + ).toEqual({ + high: { thinking: { type: "enabled", budgetTokens: 16_000 } }, + max: { thinking: { type: "enabled", budgetTokens: 31_999 } }, + }) + }) + + test("caps token budgets below the model output limit", () => { + const anthropic = target("@ai-sdk/anthropic") + anthropic.limit.output = 5_000 + expect( + ProviderTransform.reasoningVariants(model([{ type: "budget_tokens", min: 1_024, max: 64_000 }]), anthropic), + ).toEqual({ + high: { thinking: { type: "enabled", budgetTokens: 2_500 } }, + max: { thinking: { type: "enabled", budgetTokens: 4_999 } }, + }) + }) + + test("derives high and max budgets when models.dev omits max", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "budget_tokens", min: 1_024 }]), + target("@ai-sdk/anthropic", "claude-haiku-4-5"), + ), + ).toEqual({ + high: { thinking: { type: "enabled", budgetTokens: 16_000 } }, + max: { thinking: { type: "enabled", budgetTokens: 31_999 } }, + }) + }) + + test("preserves explicit inclusive budget maxima", () => { + expect( + ProviderTransform.reasoningVariants( + model([{ type: "budget_tokens", min: 1_024, max: 24_576 }]), + target("@ai-sdk/google", "gemini-2.5-pro"), + ), + ).toEqual({ + high: { thinkingConfig: { includeThoughts: true, thinkingBudget: 12_288 } }, + max: { thinkingConfig: { includeThoughts: true, thinkingBudget: 24_576 } }, + }) + }) + + test("prefers effort options over token budgets", () => { + expect( + ProviderTransform.reasoningVariants( + model([ + { type: "budget_tokens", min: 1_024, max: 64_000 }, + { type: "effort", values: ["low"] }, + ]), + target("@ai-sdk/openai"), + ), + ).toEqual({ + low: { + reasoningEffort: "low", + reasoningSummary: "auto", + include: ["reasoning.encrypted_content"], + }, + }) + }) + + test("leaves unsupported options for heuristic fallback", () => { + expect( + ProviderTransform.reasoningVariants(model([{ type: "effort", values: ["high"] }]), target("@ai-sdk/perplexity")), + ).toBeUndefined() + expect(ProviderTransform.reasoningVariants(model([{ type: "toggle" }]), target("@ai-sdk/openai"))).toBeUndefined() + }) + + test("uses model-family options for gateway and GitHub Copilot", () => { + const effort = model([{ type: "effort", values: ["high"] }]) + expect(ProviderTransform.reasoningVariants(effort, target("@ai-sdk/gateway", "anthropic/claude-sonnet-4"))).toEqual( + { + high: { thinking: { type: "adaptive", display: "summarized" }, effort: "high" }, + }, + ) + expect(ProviderTransform.reasoningVariants(effort, target("@ai-sdk/gateway", "google/gemini-3-pro"))).toEqual({ + high: { thinkingConfig: { includeThoughts: true, thinkingLevel: "high" } }, + }) + expect( + ProviderTransform.reasoningVariants(effort, target("@ai-sdk/github-copilot", "gemini-3-pro")), + ).toBeUndefined() + }) + + test.each(["@ai-sdk/cohere", "@ai-sdk/perplexity", "@ai-sdk/vercel", "@ai-sdk/alibaba", "gitlab-ai-provider"])( + "does not invent effort controls for %s", + (npm) => { + expect( + ProviderTransform.reasoningVariants(model([{ type: "effort", values: ["high"] }]), target(npm)), + ).toBeUndefined() + }, + ) +}) + describe("ProviderTransform.variants", () => { const createMockModel = (overrides: Partial = {}): any => ({ id: "test/test-model", @@ -3372,6 +3672,42 @@ describe("ProviderTransform.variants", () => { }) describe("@ai-sdk/gateway", () => { + test("configured anthropic aliases route by the API ID", () => { + const model = createMockModel({ + id: "my-claude", + providerID: "gateway", + api: { + id: "anthropic/claude-sonnet-4-6", + url: "https://gateway.ai", + npm: "@ai-sdk/gateway", + }, + }) + const result = ProviderTransform.variants(model) + expect(Object.keys(result)).toEqual(["low", "medium", "high", "max"]) + expect(result.high).toEqual({ + thinking: { + type: "adaptive", + }, + effort: "high", + }) + }) + + test("configured google aliases route by the API ID", () => { + const model = createMockModel({ + id: "my-gemini", + providerID: "gateway", + api: { + id: "google/gemini-2.5-pro", + url: "https://gateway.ai", + npm: "@ai-sdk/gateway", + }, + }) + expect(ProviderTransform.variants(model)).toEqual({ + high: { thinkingConfig: { includeThoughts: true, thinkingBudget: 16_000 } }, + max: { thinkingConfig: { includeThoughts: true, thinkingBudget: 32_768 } }, + }) + }) + test("anthropic sonnet 4.6 models return adaptive thinking options", () => { const model = createMockModel({ id: "anthropic/claude-sonnet-4-6", diff --git a/packages/opencode/test/session/llm.test.ts b/packages/opencode/test/session/llm.test.ts index 5574d29356da..61aac13ac395 100644 --- a/packages/opencode/test/session/llm.test.ts +++ b/packages/opencode/test/session/llm.test.ts @@ -1092,7 +1092,7 @@ describe("session.llm.stream", () => { const agent = { name: "test", mode: "primary", - options: {}, + options: { reasoningMode: "pro" }, permission: [{ permission: "*", pattern: "*", action: "allow" }], temperature: 0.2, } satisfies Agent.Info @@ -1123,6 +1123,7 @@ describe("session.llm.stream", () => { expect(body.model).toBe(resolved.api.id) expect(body.stream).toBe(true) expect((body.reasoning as { effort?: string } | undefined)?.effort).toBe("high") + expect((body.reasoning as { mode?: string } | undefined)?.mode).toBe("pro") const maxTokens = body.max_output_tokens as number | undefined expect(maxTokens).toBe(undefined) // match codex cli behavior diff --git a/packages/opencode/test/tool/fixtures/models-api.json b/packages/opencode/test/tool/fixtures/models-api.json index 6302a951dd89..8ead8c2a9ab7 100644 --- a/packages/opencode/test/tool/fixtures/models-api.json +++ b/packages/opencode/test/tool/fixtures/models-api.json @@ -1,709 +1,976 @@ { - "302ai": { - "id": "302ai", - "env": ["302AI_API_KEY"], + "requesty": { + "id": "requesty", + "env": ["REQUESTY_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.302.ai/v1", - "name": "302.AI", - "doc": "https://doc.302.ai", + "api": "https://router.requesty.ai/v1", + "name": "Requesty", + "doc": "https://requesty.ai/solution/llm-routing/models", "models": { - "qwen3-235b-a22b": { - "id": "qwen3-235b-a22b", - "name": "Qwen3-235B-A22B", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-29", - "last_updated": "2025-04-29", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 128000, - "output": 16384 - }, - "cost": { - "input": 0.29, - "output": 2.86 - } - }, - "grok-4.1": { - "id": "grok-4.1", - "name": "grok-4.1", + "xai/grok-4": { + "id": "xai/grok-4", + "name": "Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "knowledge": "2025-01", + "release_date": "2025-09-09", + "last_updated": "2025-09-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 256000, "output": 64000 }, "cost": { - "input": 2, - "output": 10 + "input": 3, + "output": 15, + "cache_read": 0.75, + "cache_write": 3 } }, - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax-M2", - "attachment": false, - "reasoning": false, + "xai/grok-4-fast": { + "id": "xai/grok-4-fast", + "name": "Grok 4 Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-10-26", - "last_updated": "2025-10-26", + "knowledge": "2025-01", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 2000000, + "output": 64000 }, "cost": { - "input": 0.33, - "output": 1.32 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05, + "cache_write": 0.2 } }, - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "grok-4-1-fast-reasoning", + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 1.25, + "output": 10, + "cache_read": 0.31, + "cache_write": 2.375, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "gemini-2.5-flash-nothink": { - "id": "gemini-2.5-flash-nothink", - "name": "gemini-2.5-flash-nothink", + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2025-06-24", - "last_updated": "2025-06-24", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 1048576, "output": 65536 }, "cost": { "input": 0.3, - "output": 2.5 + "output": 2.5, + "cache_read": 0.075, + "cache_write": 0.55 } }, - "grok-4.20-multi-agent-beta-0309": { - "id": "grok-4.20-multi-agent-beta-0309", - "name": "grok-4.20-multi-agent-beta-0309", + "google/gemini-3-pro-preview": { + "id": "google/gemini-3-pro-preview", + "name": "Gemini 3 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 1048576, + "output": 65536 }, "cost": { "input": 2, - "output": 6 + "output": 12, + "cache_read": 0.2, + "cache_write": 4.5 } }, - "kimi-k2-0905-preview": { - "id": "kimi-k2-0905-preview", - "name": "kimi-k2-0905-preview", - "attachment": false, - "reasoning": false, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.632, - "output": 2.53 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 1 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "claude-haiku-4-5", - "family": "claude-haiku", + "openai/gpt-5.2-chat": { + "id": "openai/gpt-5.2-chat", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-16", - "last_updated": "2025-10-16", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1, - "output": 5 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "claude-opus-4-5-20251101", - "family": "claude-opus", + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT-5.2 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25 + "input": 21, + "output": 168 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "gemini-2.5-flash-lite-preview-09-2025", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-26", - "last_updated": "2025-09-26", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "audio", "image", "video"], + "output": ["text", "audio", "image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "qwen3-235b-a22b-instruct-2507", - "attachment": false, - "reasoning": false, - "tool_call": true, + "openai/gpt-5-chat": { + "id": "openai/gpt-5-chat", + "name": "GPT-5 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.29, - "output": 1.143 + "input": 1.25, + "output": 10 } }, - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM-5V-Turbo", - "family": "glm", + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 400000, + "output": 272000 }, "cost": { - "input": 0.72, - "output": 3.2 + "input": 15, + "output": 120 } }, - "mistral-large-2512": { - "id": "mistral-large-2512", - "name": "mistral-large-2512", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4 Mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "knowledge": "2024-06", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 262144 + "context": 200000, + "output": 100000 }, "cost": { "input": 1.1, - "output": 3.3 + "output": 4.4, + "cache_read": 0.28 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "glm-4.7", - "family": "glm", - "attachment": false, + "openai/gpt-5.1-chat": { + "id": "openai/gpt-5.1-chat", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.286, - "output": 1.142 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "claude-3-5-haiku-20241022": { - "id": "claude-3-5-haiku-20241022", - "name": "claude-3-5-haiku-20241022", - "family": "claude-haiku", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.8, - "output": 4 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "doubao-seed-1-8-251215": { - "id": "doubao-seed-1-8-251215", - "name": "doubao-seed-1-8-251215", + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT-5.1-Codex-Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 224000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.114, - "output": 0.286 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "chatgpt-4o-latest": { - "id": "chatgpt-4o-latest", - "name": "chatgpt-4o-latest", + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-08-08", - "last_updated": "2024-08-08", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "output": 128000 }, "cost": { - "input": 5, - "output": 15 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "glm-5": { - "id": "glm-5", - "name": "glm-5", - "family": "glm", - "attachment": false, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.6 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "deepseek-chat": { - "id": "deepseek-chat", - "name": "Deepseek-Chat", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1-Codex-Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-11-29", - "last_updated": "2024-11-29", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 400000, + "output": 100000 }, "cost": { - "input": 0.29, - "output": 0.43 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "deepseek-v3.2-thinking": { - "id": "deepseek-v3.2-thinking", - "name": "DeepSeek-V3.2-Thinking", - "attachment": false, + "openai/gpt-5-image": { + "id": "openai/gpt-5-image", + "name": "GPT-5 Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-10-01", + "release_date": "2025-10-14", + "last_updated": "2025-10-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 400000, "output": 128000 }, "cost": { - "input": 0.29, - "output": 0.43 + "input": 5, + "output": 10, + "cache_read": 1.25 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "claude-sonnet-4-6", - "family": "claude-sonnet", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, + "structured_output": true, + "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-02-18", - "last_updated": "2026-03-13", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "gpt-5-thinking": { - "id": "gpt-5-thinking", - "name": "gpt-5-thinking", + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "glm-4.7-flashx": { - "id": "glm-4.7-flashx", - "name": "glm-4.7-flashx", - "family": "glm-flash", - "attachment": false, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-20", - "last_updated": "2026-01-20", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.0715, - "output": 0.429 + "input": 0.25, + "output": 2, + "cache_read": 0.03 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "gemini-3-flash-preview", + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "qwen-plus": { - "id": "qwen-plus", - "name": "Qwen-Plus", - "family": "qwen", - "attachment": false, - "reasoning": false, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 16000, + "output": 4000 }, "cost": { - "input": 0.12, - "output": 1.2 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01 } }, - "grok-4.20-beta-0309-non-reasoning": { - "id": "grok-4.20-beta-0309-non-reasoning", - "name": "grok-4.20-beta-0309-non-reasoning", + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6 + "input": 30, + "output": 180, + "cache_read": 30 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "claude-opus-4-7", - "family": "claude-opus", + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2026-01-31", - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "knowledge": "2024-10", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 0.15, + "output": 0.6, + "cache_read": 0.08 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "gpt-5-mini", - "family": "gpt-mini", + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "temperature": true, + "knowledge": "2024-10-01", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -711,96 +978,146 @@ "open_weights": false, "limit": { "context": 400000, - "input": 272000, "output": 128000 }, "cost": { - "input": 0.25, - "output": 2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "gemini-3-pro-preview", + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "knowledge": "2025-08-31", + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 2, - "output": 12 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "MiniMax-M2.7": { - "id": "MiniMax-M2.7", - "name": "MiniMax-M2.7", - "attachment": false, - "reasoning": false, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "qwen3-max-2025-09-23": { - "id": "qwen3-max-2025-09-23", - "name": "qwen3-max-2025-09-23", - "attachment": false, - "reasoning": false, + "anthropic/claude-3-7-sonnet": { + "id": "anthropic/claude-3-7-sonnet", + "name": "Claude Sonnet 3.7", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "knowledge": "2024-01", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 258048, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.86, - "output": 3.43 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "claude-sonnet-4-5-20250929", + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -812,92 +1129,143 @@ }, "cost": { "input": 3, - "output": 15 + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "qwen-flash": { - "id": "qwen-flash", - "name": "Qwen-Flash", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-4-5": { + "id": "anthropic/claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.022, - "output": 0.22 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "gemini-2.5-pro", - "family": "gemini-pro", + "anthropic/claude-sonnet-4-5": { + "id": "anthropic/claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 65536 + "output": 64000 }, "cost": { - "input": 1.25, - "output": 10 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "grok-4-1-fast-non-reasoning", - "attachment": true, - "reasoning": false, + "anthropic/claude-opus-4-1": { + "id": "anthropic/claude-opus-4-1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "claude-3-5-haiku-latest": { - "id": "claude-3-5-haiku-latest", - "name": "claude-3-5-haiku-latest", + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", "family": "claude-haiku", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "knowledge": "2025-02-01", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -905,477 +1273,580 @@ "open_weights": false, "limit": { "context": 200000, - "output": 8192 + "output": 62000 }, "cost": { - "input": 0.8, - "output": 4 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "claude-opus-4-5-20251101-thinking": { - "id": "claude-opus-4-5-20251101-thinking", - "name": "claude-opus-4-5-20251101-thinking", + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { "input": 5, - "output": 25 + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "gpt-5.2", - "family": "gpt", + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, + "temperature": true, "knowledge": "2025-08-31", - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 1.75, - "output": 14 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "gpt-5.4-mini", - "family": "gpt-mini", + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.75, - "output": 4.5 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } - }, - "gemini-3-pro-image-preview": { - "id": "gemini-3-pro-image-preview", - "name": "gemini-3-pro-image-preview", - "attachment": true, - "reasoning": false, - "tool_call": false, + } + } + }, + "qiniu-ai": { + "id": "qiniu-ai", + "env": ["QINIU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.qnaigc.com/v1", + "name": "Qiniu", + "doc": "https://developer.qiniu.com/aitokenapi", + "models": { + "deepseek-r1-0528": { + "id": "deepseek-r1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 64000 - }, - "cost": { - "input": 2, - "output": 120 + "context": 128000, + "output": 32000 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "glm-5.1", - "family": "glm", + "doubao-1.5-thinking-pro": { + "id": "doubao-1.5-thinking-pro", + "name": "Doubao 1.5 Thinking Pro", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-04-10", - "last_updated": "2026-04-10", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 0.86, - "output": 3.5 + "context": 128000, + "output": 16000 } }, - "qwen-max-latest": { - "id": "qwen-max-latest", - "name": "Qwen-Max-Latest", - "family": "qwen", - "attachment": false, + "qwen3-vl-30b-a3b-thinking": { + "id": "qwen3-vl-30b-a3b-thinking", + "name": "Qwen3-Vl 30b A3b Thinking", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-04-03", - "last_updated": "2025-01-25", + "release_date": "2026-02-09", + "last_updated": "2026-02-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 - }, - "cost": { - "input": 0.343, - "output": 1.372 + "context": 128000, + "output": 32000 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "gpt-5.4-nano", - "family": "gpt-nano", + "claude-3.5-haiku": { + "id": "claude-3.5-haiku", + "name": "Claude 3.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "cost": { - "input": 0.2, - "output": 1.25 + "context": 200000, + "output": 8192 } }, - "gemini-2.5-flash-image": { - "id": "gemini-2.5-flash-image", - "name": "gemini-2.5-flash-image", - "attachment": true, + "deepseek-v3-0324": { + "id": "deepseek-v3-0324", + "name": "DeepSeek-V3-0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-10-08", - "last_updated": "2025-10-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 - }, - "cost": { - "input": 0.3, - "output": 30 + "context": 128000, + "output": 16000 } }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM-4.5", - "family": "glm", + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235b A22B Instruct 2507", + "description": "Tool-capable chat model for instruction following and agentic application workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2025-08-12", + "last_updated": "2025-08-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 - }, - "cost": { - "input": 0.286, - "output": 1.142 + "context": 262144, + "output": 64000 } }, - "gpt-5.4-mini-2026-03-17": { - "id": "gpt-5.4-mini-2026-03-17", - "name": "gpt-5.4-mini-2026-03-17", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "deepseek-v3": { + "id": "deepseek-v3", + "name": "DeepSeek-V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-13", + "last_updated": "2025-08-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "cost": { - "input": 0.75, - "output": 4.5 + "context": 128000, + "output": 16000 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "gemini-2.5-flash", - "family": "gemini-flash", - "attachment": true, + "kimi-k2": { + "id": "kimi-k2", + "name": "Kimi K2", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 - }, - "cost": { - "input": 0.3, - "output": 2.5 + "context": 128000, + "output": 128000 } }, - "gpt-5.2-chat-latest": { - "id": "gpt-5.2-chat-latest", - "name": "gpt-5.2-chat-latest", - "family": "gpt-codex", - "attachment": true, + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 - }, - "cost": { - "input": 1.75, - "output": 14 + "context": 40000, + "output": 4096 } }, - "doubao-seed-1-6-vision-250815": { - "id": "doubao-seed-1-6-vision-250815", - "name": "doubao-seed-1-6-vision-250815", - "attachment": true, + "qwen3-max-preview": { + "id": "qwen3-max-preview", + "name": "Qwen3 Max Preview", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-09-06", + "last_updated": "2025-09-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "output": 32000 - }, - "cost": { - "input": 0.114, - "output": 1.143 + "output": 64000 } }, - "gemini-3.1-flash-image-preview": { - "id": "gemini-3.1-flash-image-preview", - "name": "gemini-3.1-flash-image-preview", + "claude-3.5-sonnet": { + "id": "claude-3.5-sonnet", + "name": "Claude 3.5 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-27", - "last_updated": "2026-02-27", + "release_date": "2025-09-09", + "last_updated": "2025-09-09", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - }, - "cost": { - "input": 0.5, - "output": 60 + "context": 200000, + "output": 8200 } }, - "MiniMax-M2.7-highspeed": { - "id": "MiniMax-M2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", + "qwen3-next-80b-a3b-instruct": { + "id": "qwen3-next-80b-a3b-instruct", + "name": "Qwen3 Next 80B A3B Instruct", + "description": "Tool-capable chat model for instruction following and agentic application workflows", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "release_date": "2025-09-12", + "last_updated": "2025-09-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 32768 + } + }, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] }, - "cost": { - "input": 0.6, - "output": 4.8 + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 } }, - "glm-4.5-x": { - "id": "glm-4.5-x", - "name": "glm-4.5-x", - "family": "glm", - "attachment": false, - "reasoning": false, + "claude-4.5-haiku": { + "id": "claude-4.5-haiku", + "name": "Claude 4.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2025-10-16", + "last_updated": "2025-10-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 + } + }, + "kling-v2-6": { + "id": "kling-v2-6", + "name": "Kling-V2 6", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2026-01-13", + "last_updated": "2026-01-13", + "modalities": { + "input": ["text", "image", "video"], + "output": ["video"] }, - "cost": { - "input": 1.143, - "output": 2.29 + "open_weights": false, + "limit": { + "context": 99999999, + "output": 99999999 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", - "name": "MiniMax-M2.1", + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM 4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 131072 - }, - "cost": { - "input": 0.3, - "output": 1.2 + "context": 131072, + "output": 98304 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "gpt-5.1", - "family": "gpt", + "claude-4.1-opus": { + "id": "claude-4.1-opus", + "name": "Claude 4.1 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "output": 32000 + } + }, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] }, - "cost": { - "input": 1.25, - "output": 10 + "open_weights": false, + "limit": { + "context": 1048576, + "output": 64000 } }, - "kimi-k2-thinking-turbo": { - "id": "kimi-k2-thinking-turbo", - "name": "kimi-k2-thinking-turbo", + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { "input": ["text"], "output": ["text"] @@ -1383,193 +1854,179 @@ "open_weights": false, "limit": { "context": 262144, - "output": 262144 - }, - "cost": { - "input": 1.265, - "output": 9.119 + "output": 65536 } }, - "deepseek-reasoner": { - "id": "deepseek-reasoner", - "name": "Deepseek-Reasoner", - "family": "deepseek-thinking", - "attachment": false, + "doubao-seed-2.0-pro": { + "id": "doubao-seed-2.0-pro", + "name": "Doubao Seed 2.0 Pro", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 256000, "output": 128000 - }, - "cost": { - "input": 0.29, - "output": 0.43 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "grok-4-fast-reasoning", + "doubao-seed-1.6": { + "id": "doubao-seed-1.6", + "name": "Doubao-Seed 1.6", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "release_date": "2025-08-15", + "last_updated": "2025-08-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 - }, - "cost": { - "input": 0.2, - "output": 0.5 + "context": 256000, + "output": 32000 } }, - "claude-opus-4-1-20250805-thinking": { - "id": "claude-opus-4-1-20250805-thinking", - "name": "claude-opus-4-1-20250805-thinking", + "doubao-seed-1.6-thinking": { + "id": "doubao-seed-1.6-thinking", + "name": "Doubao-Seed 1.6 Thinking", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-05-27", - "last_updated": "2025-05-27", + "release_date": "2025-08-15", + "last_updated": "2025-08-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 256000, "output": 32000 + } + }, + "gemini-2.0-flash": { + "id": "gemini-2.0-flash", + "name": "Gemini 2.0 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] }, - "cost": { - "input": 15, - "output": 75 + "open_weights": false, + "limit": { + "context": 1048576, + "output": 8192 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "glm-4.5-air", - "family": "glm-air", + "qwen-max-2025-01-25": { + "id": "qwen-max-2025-01-25", + "name": "Qwen2.5-Max-2025-01-25", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 - }, - "cost": { - "input": 0.1143, - "output": 0.286 + "context": 128000, + "output": 4096 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "gpt-5.4-pro", - "family": "gpt-pro", + "claude-4.0-sonnet": { + "id": "claude-4.0-sonnet", + "name": "Claude 4.0 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "cost": { - "input": 30, - "output": 180, - "cache_read": 0, - "cache_write": 0, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 60, - "output": 270 - } + "context": 200000, + "output": 64000 } }, - "glm-5-turbo": { - "id": "glm-5-turbo", - "name": "glm-5-turbo", - "family": "glm", + "doubao-1.5-pro-32k": { + "id": "doubao-1.5-pro-32k", + "name": "Doubao 1.5 Pro 32k", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 0.72, - "output": 3.2 + "context": 128000, + "output": 12000 } }, - "qwen3-30b-a3b": { - "id": "qwen3-30b-a3b", - "name": "Qwen3-30B-A3B", - "family": "qwen", + "qwen3-30b-a3b-instruct-2507": { + "id": "qwen3-30b-a3b-instruct-2507", + "name": "Qwen3 30b A3b Instruct 2507", + "description": "Tool-capable chat model for instruction following and agentic application workflows", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "release_date": "2026-02-04", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] @@ -1577,1471 +2034,1552 @@ "open_weights": false, "limit": { "context": 128000, - "output": 8192 - }, - "cost": { - "input": 0.11, - "output": 1.08 + "output": 32000 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "claude-opus-4-5", - "family": "claude-opus", - "attachment": true, + "qwen3-next-80b-a3b-thinking": { + "id": "qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B Thinking", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "release_date": "2025-09-12", + "last_updated": "2025-09-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 5, - "output": 25 + "context": 131072, + "output": 32768 } }, - "glm-4.5v": { - "id": "glm-4.5v", - "name": "GLM-4.5V", - "family": "glm", - "attachment": true, + "qwen3-235b-a22b-thinking-2507": { + "id": "qwen3-235b-a22b-thinking-2507", + "name": "Qwen3 235B A22B Thinking 2507", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", "release_date": "2025-08-12", "last_updated": "2025-08-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 16384 - }, - "cost": { - "input": 0.29, - "output": 0.86 + "context": 262144, + "output": 4096 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "glm-4.6", - "family": "glm", + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 0.286, - "output": 1.142 + "context": 128000, + "output": 32000 } }, - "claude-opus-4-6-thinking": { - "id": "claude-opus-4-6-thinking", - "name": "claude-opus-4-6-thinking", + "doubao-1.5-vision-pro": { + "id": "doubao-1.5-vision-pro", + "name": "Doubao 1.5 Vision Pro", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-02-06", - "last_updated": "2026-03-13", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 5, - "output": 25 + "context": 128000, + "output": 16000 } }, - "gemini-2.5-flash-preview-09-2025": { - "id": "gemini-2.5-flash-preview-09-2025", - "name": "gemini-2.5-flash-preview-09-2025", + "gemini-3.0-pro-image-preview": { + "id": "gemini-3.0-pro-image-preview", + "name": "Gemini 3.0 Pro Image Preview", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-26", - "last_updated": "2025-09-26", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 - }, - "cost": { - "input": 0.3, - "output": 2.5 + "context": 32768, + "output": 8192 } }, - "claude-sonnet-4-6-thinking": { - "id": "claude-sonnet-4-6-thinking", - "name": "claude-sonnet-4-6-thinking", + "gemini-2.5-flash-image": { + "id": "gemini-2.5-flash-image", + "name": "Gemini 2.5 Flash Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-08", - "release_date": "2026-02-18", - "last_updated": "2026-03-13", + "release_date": "2025-10-22", + "last_updated": "2025-10-22", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 + "context": 32768, + "output": 8192 } }, - "glm-4.6v": { - "id": "glm-4.6v", - "name": "GLM-4.6V", - "family": "glm", + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 - }, - "cost": { - "input": 0.145, - "output": 0.43 + "context": 1048576, + "output": 64000 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "claude-opus-4-1-20250805", - "family": "claude-opus", + "claude-3.7-sonnet": { + "id": "claude-3.7-sonnet", + "name": "Claude 3.7 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", "release_date": "2025-08-05", "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 32000 - }, - "cost": { - "input": 15, - "output": 75 + "output": 128000 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "gpt-5.4", - "family": "gpt", - "attachment": true, + "qwen3-30b-a3b-thinking-2507": { + "id": "qwen3-30b-a3b-thinking-2507", + "name": "Qwen3 30b A3b Thinking 2507", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": false, + "temperature": true, + "release_date": "2026-02-04", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "cache_write": 0, - "tiers": [ - { - "input": 5, - "output": 22.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 5, - "output": 22.5 - } + "context": 126000, + "output": 32000 } }, - "gpt-5.1-chat-latest": { - "id": "gpt-5.1-chat-latest", - "name": "gpt-5.1-chat-latest", - "family": "gpt-codex", + "qwen2.5-vl-72b-instruct": { + "id": "qwen2.5-vl-72b-instruct", + "name": "Qwen 2.5 VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 16384 - }, - "cost": { - "input": 1.25, - "output": 10 + "output": 8192 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "claude-haiku-4-5-20251001", - "family": "claude-haiku", - "attachment": true, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-16", - "last_updated": "2025-10-16", + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 1, - "output": 5 + "context": 128000, + "output": 4096 } }, - "MiniMax-M1": { - "id": "MiniMax-M1", - "name": "MiniMax-M1", - "family": "minimax", - "attachment": false, - "reasoning": false, + "doubao-seed-1.6-flash": { + "id": "doubao-seed-1.6-flash", + "name": "Doubao-Seed 1.6 Flash", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-06-16", - "last_updated": "2025-06-16", + "release_date": "2025-08-15", + "last_updated": "2025-08-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 0.132, - "output": 1.254 + "context": 256000, + "output": 32000 } }, - "gpt-5.4-nano-2026-03-17": { - "id": "gpt-5.4-nano-2026-03-17", - "name": "gpt-5.4-nano-2026-03-17", - "family": "gpt-nano", - "attachment": true, + "deepseek-v3.1": { + "id": "deepseek-v3.1", + "name": "DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-19", - "last_updated": "2026-03-19", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "cost": { - "input": 0.2, - "output": 1.25 + "context": 128000, + "output": 32000 } }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "claude-sonnet-4-20250514", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "qwen3-235b-a22b": { + "id": "qwen3-235b-a22b", + "name": "Qwen 3 235B A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 + "context": 128000, + "output": 32000 } }, "qwen3-coder-480b-a35b-instruct": { "id": "qwen3-coder-480b-a35b-instruct", - "name": "qwen3-coder-480b-a35b-instruct", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2025-08-14", + "last_updated": "2025-08-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.86, - "output": 3.43 + "context": 262000, + "output": 4096 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "claude-opus-4-6", - "family": "claude-opus", + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-06", - "last_updated": "2026-03-13", + "release_date": "2026-02-22", + "last_updated": "2026-02-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 5, - "output": 25 + "context": 256000, + "output": 64000 } }, - "doubao-seed-code-preview-251028": { - "id": "doubao-seed-code-preview-251028", - "name": "doubao-seed-code-preview-251028", - "attachment": true, - "reasoning": false, + "mimo-v2-flash": { + "id": "mimo-v2-flash", + "name": "Mimo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-11-11", - "last_updated": "2025-11-11", + "knowledge": "2024-12-01", + "release_date": "2025-12-16", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 256000, - "output": 32000 + "output": 256000 }, "cost": { - "input": 0.17, - "output": 1.14 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "gpt-4.1-nano", - "family": "gpt-nano", + "qwen-vl-max-2025-01-25": { + "id": "qwen-vl-max-2025-01-25", + "name": "Qwen VL-MAX-2025-01-25", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 - }, - "cost": { - "input": 0.1, - "output": 0.4 + "context": 128000, + "output": 4096 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "deepseek-v3.2", - "attachment": false, + "qwen2.5-vl-7b-instruct": { + "id": "qwen2.5-vl-7b-instruct", + "name": "Qwen 2.5 VL 7B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, "output": 8192 - }, - "cost": { - "input": 0.29, - "output": 0.43 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "gpt-5-pro", - "family": "gpt-pro", - "attachment": true, + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-08", - "last_updated": "2025-10-08", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 272000 - }, - "cost": { - "input": 15, - "output": 120 + "context": 131000, + "output": 4096 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "gpt-4o", - "family": "gpt", + "claude-4.5-opus": { + "id": "claude-4.5-opus", + "name": "Claude 4.5 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "release_date": "2025-11-25", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 - }, - "cost": { - "input": 2.5, - "output": 10 + "context": 200000, + "output": 200000 } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "claude-sonnet-4-5", - "family": "claude-sonnet", + "gemini-3.0-pro-preview": { + "id": "gemini-3.0-pro-preview", + "name": "Gemini 3.0 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "pdf", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 1000000, "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 } }, - "gpt-5": { - "id": "gpt-5", - "name": "gpt-5", - "family": "gpt", + "doubao-seed-2.0-mini": { + "id": "doubao-seed-2.0-mini", + "name": "Doubao Seed 2.0 Mini", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "structured_output": false, + "temperature": true, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "cost": { - "input": 1.25, - "output": 10 + "context": 256000, + "output": 32000 } }, - "grok-4.20-beta-0309-reasoning": { - "id": "grok-4.20-beta-0309-reasoning", - "name": "grok-4.20-beta-0309-reasoning", + "claude-4.0-opus": { + "id": "claude-4.0-opus", + "name": "Claude 4.0 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "output": 32000 + } + }, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", + "modalities": { + "input": ["text"], + "output": ["text"] }, - "cost": { - "input": 2, - "output": 6 + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "claude-opus-4-20250514", - "family": "claude-opus", + "gemini-3.0-flash-preview": { + "id": "gemini-3.0-flash-preview", + "name": "Gemini 3.0 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 - }, - "cost": { - "input": 15, - "output": 75 + "context": 1000000, + "output": 64000 } }, - "glm-for-coding": { - "id": "glm-for-coding", - "name": "glm-for-coding", - "family": "glm", + "MiniMax-M1": { + "id": "MiniMax-M1", + "name": "MiniMax M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 0.086, - "output": 0.343 + "context": 1000000, + "output": 80000 } }, - "claude-sonnet-4-5-20250929-thinking": { - "id": "claude-sonnet-4-5-20250929-thinking", - "name": "claude-sonnet-4-5-20250929-thinking", - "attachment": true, + "qwen-turbo": { + "id": "qwen-turbo", + "name": "Qwen-Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 + "context": 1000000, + "output": 4096 } }, - "glm-4.5-airx": { - "id": "glm-4.5-airx", - "name": "glm-4.5-airx", - "family": "glm", + "qwen3-30b-a3b": { + "id": "qwen3-30b-a3b", + "name": "Qwen3 30B A3B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 - }, - "cost": { - "input": 0.572, - "output": 1.714 + "context": 40000, + "output": 4096 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "gpt-4.1", - "family": "gpt", + "claude-4.5-sonnet": { + "id": "claude-4.5-sonnet", + "name": "Claude 4.5 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 - }, - "cost": { - "input": 2, - "output": 8 + "context": 200000, + "output": 64000 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "kimi-k2-thinking", - "attachment": false, + "doubao-seed-2.0-lite": { + "id": "doubao-seed-2.0-lite", + "name": "Doubao Seed 2.0 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - }, - "cost": { - "input": 0.575, - "output": 2.3 + "context": 256000, + "output": 32000 } }, "gemini-2.0-flash-lite": { "id": "gemini-2.0-flash-lite", - "name": "gemini-2.0-flash-lite", - "family": "gemini-flash-lite", + "name": "Gemini 2.0 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-06-16", - "last_updated": "2025-06-16", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, + "context": 1048576, "output": 8192 - }, - "cost": { - "input": 0.075, - "output": 0.3 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "gpt-4.1-mini", - "family": "gpt-mini", + "doubao-seed-2.0-code": { + "id": "doubao-seed-2.0-code", + "name": "Doubao Seed 2.0 Code", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 - }, - "cost": { - "input": 0.4, - "output": 1.6 + "context": 256000, + "output": 128000 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "grok-4-fast-non-reasoning", - "attachment": true, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "release_date": "2025-11-07", + "last_updated": "2025-11-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 - }, - "cost": { - "input": 0.2, - "output": 0.5 + "context": 256000, + "output": 100000 } }, - "doubao-seed-1-6-thinking-250715": { - "id": "doubao-seed-1-6-thinking-250715", - "name": "doubao-seed-1-6-thinking-250715", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Moonshotai/Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-07-15", - "last_updated": "2025-07-15", + "release_date": "2026-01-28", + "last_updated": "2026-01-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "output": 16000 + "output": 256000 + } + }, + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-09-08", + "last_updated": "2025-09-08", + "modalities": { + "input": ["text"], + "output": ["text"] }, - "cost": { - "input": 0.121, - "output": 1.21 + "open_weights": false, + "limit": { + "context": 256000, + "output": 100000 } }, - "ministral-14b-2512": { - "id": "ministral-14b-2512", - "name": "ministral-14b-2512", + "stepfun-ai/gelab-zero-4b-preview": { + "id": "stepfun-ai/gelab-zero-4b-preview", + "name": "Stepfun-Ai/Gelab Zero 4b Preview", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 - }, - "cost": { - "input": 0.33, - "output": 0.33 + "context": 8192, + "output": 4096 } - } - } - }, - "alibaba": { - "id": "alibaba", - "env": ["DASHSCOPE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1", - "name": "Alibaba", - "doc": "https://www.alibabacloud.com/help/en/model-studio/models", - "models": { - "qwen3-235b-a22b": { - "id": "qwen3-235b-a22b", - "name": "Qwen3 235B-A22B", - "family": "qwen", + }, + "x-ai/grok-code-fast-1": { + "id": "x-ai/grok-code-fast-1", + "name": "x-AI/Grok-Code-Fast 1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-09-02", + "last_updated": "2025-09-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 - }, - "cost": { - "input": 0.7, - "output": 2.8, - "reasoning": 8.4 + "context": 256000, + "output": 10000 } }, - "qwen3.5-122b-a10b": { - "id": "qwen3.5-122b-a10b", - "name": "Qwen3.5 122B-A10B", - "family": "qwen", + "x-ai/grok-4.1-fast-reasoning": { + "id": "x-ai/grok-4.1-fast-reasoning", + "name": "X-Ai/Grok 4.1 Fast Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-02-23", - "last_updated": "2026-02-23", + "release_date": "2025-12-19", + "last_updated": "2025-12-19", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.4, - "output": 3.2 + "context": 20000000, + "output": 2000000 } }, - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3 Coder Plus", - "family": "qwen", - "attachment": false, + "x-ai/grok-4.1-fast-non-reasoning": { + "id": "x-ai/grok-4.1-fast-non-reasoning", + "name": "X-Ai/Grok 4.1 Fast Non Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2025-12-19", + "last_updated": "2025-12-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 - }, - "cost": { - "input": 1, - "output": 5 + "context": 2000000, + "output": 2000000 } }, - "qwen3.6-27b": { - "id": "qwen3.6-27b", - "name": "Qwen3.6 27B", - "family": "qwen", + "x-ai/grok-4-fast-reasoning": { + "id": "x-ai/grok-4-fast-reasoning", + "name": "X-Ai/Grok-4-Fast-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.6, - "output": 3.6 + "context": 2000000, + "output": 2000000 } }, - "qwen3.5-27b": { - "id": "qwen3.5-27b", - "name": "Qwen3.5 27B", - "family": "qwen", + "x-ai/grok-4-fast": { + "id": "x-ai/grok-4-fast", + "name": "x-AI/Grok-4-Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-02-23", - "last_updated": "2026-02-23", + "release_date": "2025-09-20", + "last_updated": "2025-09-20", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.3, - "output": 2.4 + "context": 2000000, + "output": 2000000 } }, - "qwen-vl-ocr": { - "id": "qwen-vl-ocr", - "name": "Qwen-VL OCR", - "family": "qwen", - "attachment": false, + "x-ai/grok-4-fast-non-reasoning": { + "id": "x-ai/grok-4-fast-non-reasoning", + "name": "X-Ai/Grok-4-Fast-Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-10-28", - "last_updated": "2025-04-13", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 34096, - "output": 4096 - }, - "cost": { - "input": 0.72, - "output": 0.72 + "context": 2000000, + "output": 2000000 } }, - "qwen-omni-turbo-realtime": { - "id": "qwen-omni-turbo-realtime", - "name": "Qwen-Omni Turbo Realtime", - "family": "qwen", + "x-ai/grok-4.1-fast": { + "id": "x-ai/grok-4.1-fast", + "name": "x-AI/Grok-4.1-Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 2048 + "context": 2000000, + "output": 2000000 + } + }, + "z-ai/autoglm-phone-9b": { + "id": "z-ai/autoglm-phone-9b", + "name": "Z-Ai/Autoglm Phone 9b", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "cost": { - "input": 0.27, - "output": 1.07, - "input_audio": 4.44, - "output_audio": 8.89 + "open_weights": false, + "limit": { + "context": 12800, + "output": 4096 } }, - "qwen3-8b": { - "id": "qwen3-8b", - "name": "Qwen3 8B", - "family": "qwen", + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "Z-Ai/GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 - }, - "cost": { - "input": 0.18, - "output": 0.7, - "reasoning": 2.1 + "context": 200000, + "output": 200000 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen3.5 397B-A17B", - "family": "qwen", - "attachment": true, - "reasoning": true, + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "Z-AI/GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "release_date": "2025-10-11", + "last_updated": "2025-10-11", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.6, - "output": 3.6 + "context": 200000, + "output": 200000 } }, - "qwq-plus": { - "id": "qwq-plus", - "name": "QwQ Plus", - "family": "qwen", + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "Z-Ai/GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 - }, - "cost": { - "input": 0.8, - "output": 2.4 + "context": 200000, + "output": 128000 } }, - "qwen-vl-plus": { - "id": "qwen-vl-plus", - "name": "Qwen-VL Plus", - "family": "qwen", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "OpenAI/GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-08-15", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 - }, - "cost": { - "input": 0.21, - "output": 0.63 + "context": 400000, + "output": 128000 } }, - "qwen3-livetranslate-flash-realtime": { - "id": "qwen3-livetranslate-flash-realtime", - "name": "Qwen3-LiveTranslate Flash Realtime", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "OpenAI/GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 53248, - "output": 4096 - }, - "cost": { - "input": 10, - "output": 10, - "input_audio": 10, - "output_audio": 38 + "context": 400000, + "output": 128000 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", - "family": "qwen", + "xiaomi/mimo-v2-flash": { + "id": "xiaomi/mimo-v2-flash", + "name": "Xiaomi/Mimo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-12-01", + "release_date": "2025-12-16", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.7, - "output": 2.8, - "reasoning": 8.4 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01 } }, - "qwen-max": { - "id": "qwen-max", - "name": "Qwen Max", - "family": "qwen", + "stepfun/step-3.5-flash": { + "id": "stepfun/step-3.5-flash", + "name": "Stepfun/Step-3.5 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2026-02-02", + "last_updated": "2026-02-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 64000, + "output": 4096 + } + }, + "meituan/longcat-flash-lite": { + "id": "meituan/longcat-flash-lite", + "name": "Meituan/Longcat-Flash-Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-03", - "last_updated": "2025-01-25", + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 256000, + "output": 320000 + } + }, + "meituan/longcat-flash-chat": { + "id": "meituan/longcat-flash-chat", + "name": "Meituan/Longcat-Flash-Chat", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-11-05", + "last_updated": "2025-11-05", + "modalities": { + "input": ["text"], + "output": ["text"] }, - "cost": { - "input": 1.6, - "output": 6.4 + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 } }, - "qwen-plus": { - "id": "qwen-plus", - "name": "Qwen Plus", - "family": "qwen", + "deepseek/deepseek-v3.1-terminus": { + "id": "deepseek/deepseek-v3.1-terminus", + "name": "DeepSeek/DeepSeek-V3.1-Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-09-11", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 - }, - "cost": { - "input": 0.4, - "output": 1.2, - "reasoning": 4 + "context": 128000, + "output": 32000 } }, - "qwen3.6-35b-a3b": { - "id": "qwen3.6-35b-a3b", - "name": "Qwen3.6 35B-A3B", - "family": "qwen", - "attachment": true, + "deepseek/deepseek-v3.1-terminus-thinking": { + "id": "deepseek/deepseek-v3.1-terminus-thinking", + "name": "DeepSeek/DeepSeek-V3.1-Terminus-Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 32000 + } + }, + "deepseek/deepseek-v3.2-exp-thinking": { + "id": "deepseek/deepseek-v3.2-exp-thinking", + "name": "DeepSeek/DeepSeek-V3.2-Exp-Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text"], + "output": ["text"] }, - "cost": { - "input": 0.248, - "output": 1.485 + "open_weights": false, + "limit": { + "context": 128000, + "output": 32000 } }, - "qwen-omni-turbo": { - "id": "qwen-omni-turbo", - "name": "Qwen-Omni Turbo", - "family": "qwen", + "deepseek/deepseek-v3.2-exp": { + "id": "deepseek/deepseek-v3.2-exp", + "name": "DeepSeek/DeepSeek-V3.2-Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01-19", - "last_updated": "2025-03-26", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 2048 - }, - "cost": { - "input": 0.07, - "output": 0.27, - "input_audio": 4.44, - "output_audio": 8.89 + "context": 128000, + "output": 32000 } }, - "qwen-flash": { - "id": "qwen-flash", - "name": "Qwen Flash", - "family": "qwen", + "deepseek/deepseek-v3.2-251201": { + "id": "deepseek/deepseek-v3.2-251201", + "name": "Deepseek/DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 128000, + "output": 32000 + } + }, + "deepseek/deepseek-math-v2": { + "id": "deepseek/deepseek-math-v2", + "name": "Deepseek/Deepseek-Math-V2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-12-04", + "last_updated": "2025-12-04", + "modalities": { + "input": ["text"], + "output": ["text"] }, - "cost": { - "input": 0.05, - "output": 0.4 + "open_weights": false, + "limit": { + "context": 160000, + "output": 160000 } }, - "qwen2-5-vl-7b-instruct": { - "id": "qwen2-5-vl-7b-instruct", - "name": "Qwen2.5-VL 7B Instruct", - "family": "qwen", + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "Minimax/Minimax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 - }, - "cost": { - "input": 0.35, - "output": 1.05 + "context": 204800, + "output": 128000 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen", + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "Minimax/Minimax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 - }, - "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5 + "context": 204800, + "output": 128000 + } + }, + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "Minimax/Minimax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" } + ], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-10-28", + "last_updated": "2025-10-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3 Max", + "minimax/minimax-m2.5-highspeed": { + "id": "minimax/minimax-m2.5-highspeed", + "name": "Minimax/Minimax-M2.5 Highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 128000 + } + } + } + }, + "alibaba-cn": { + "id": "alibaba-cn", + "env": ["DASHSCOPE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://dashscope.aliyuncs.com/compatible-mode/v1", + "name": "Alibaba (China)", + "doc": "https://www.alibabacloud.com/help/en/model-studio/models", + "models": { + "qwen2-5-math-72b-instruct": { + "id": "qwen2-5-math-72b-instruct", + "name": "Qwen2.5-Math 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 4096, + "output": 3072 + }, + "cost": { + "input": 0.574, + "output": 1.721 + } + }, + "deepseek-r1-0528": { + "id": "deepseek-r1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.2, - "output": 6 + "input": 0.574, + "output": 2.294 } }, "qwen3-omni-flash": { "id": "qwen3-omni-flash", "name": "Qwen3-Omni Flash", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, "knowledge": "2024-04", @@ -3057,99 +3595,126 @@ "output": 16384 }, "cost": { - "input": 0.43, - "output": 1.66, - "input_audio": 3.81, - "output_audio": 15.11 + "input": 0.058, + "output": 0.23, + "input_audio": 3.584, + "output_audio": 7.168 } }, - "qwen2-5-72b-instruct": { - "id": "qwen2-5-72b-instruct", - "name": "Qwen2.5 72B Instruct", - "family": "qwen", + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 1.4, - "output": 5.6 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "qwen3-vl-235b-a22b": { - "id": "qwen3-vl-235b-a22b", - "name": "Qwen3-VL 235B-A22B", + "qwen-plus": { + "id": "qwen-plus", + "name": "Qwen Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-09-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 1000000, "output": 32768 }, "cost": { - "input": 0.7, - "output": 2.8, - "reasoning": 8.4 + "input": 0.115, + "output": 0.287, + "reasoning": 1.147 } }, - "qwen3-asr-flash": { - "id": "qwen3-asr-flash", - "name": "Qwen3-ASR Flash", + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-09-08", - "last_updated": "2025-09-08", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 53248, - "output": 4096 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.035, - "output": 0.035 + "input": 0.216, + "output": 0.861 } }, - "qwen3-next-80b-a3b-thinking": { - "id": "qwen3-next-80b-a3b-thinking", - "name": "Qwen3-Next 80B-A3B (Thinking)", + "qwen2-5-coder-7b-instruct": { + "id": "qwen2-5-coder-7b-instruct", + "name": "Qwen2.5-Coder 7B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09", - "last_updated": "2025-09", + "knowledge": "2024-04", + "release_date": "2024-11", + "last_updated": "2024-11", "modalities": { "input": ["text"], "output": ["text"] @@ -3157,99 +3722,113 @@ "open_weights": true, "limit": { "context": 131072, - "output": 32768 + "output": 8192 }, "cost": { - "input": 0.5, - "output": 6 + "input": 0.144, + "output": 0.287 } }, - "qwen-mt-plus": { - "id": "qwen-mt-plus", - "name": "Qwen-MT Plus", - "family": "qwen", + "deepseek-v3": { + "id": "deepseek-v3", + "name": "DeepSeek V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01", - "last_updated": "2025-01", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, + "context": 65536, "output": 8192 }, "cost": { - "input": 2.46, - "output": 7.37 + "input": 0.287, + "output": 1.147 } }, - "qwen-vl-max": { - "id": "qwen-vl-max", - "name": "Qwen-VL Max", + "qwen3-omni-flash-realtime": { + "id": "qwen3-omni-flash-realtime", + "name": "Qwen3-Omni Flash Realtime", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2024-04-08", - "last_updated": "2025-08-13", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.8, - "output": 3.2 + "input": 0.23, + "output": 0.918, + "input_audio": 3.584, + "output_audio": 7.168 } }, - "qwen3-coder-flash": { - "id": "qwen3-coder-flash", - "name": "Qwen3 Coder Flash", - "family": "qwen", + "deepseek-r1-distill-llama-70b": { + "id": "deepseek-r1-distill-llama-70b", + "name": "DeepSeek R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.287, + "output": 0.861 } }, - "qwen2-5-7b-instruct": { - "id": "qwen2-5-7b-instruct", - "name": "Qwen2.5 7B Instruct", + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 38912 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] @@ -3257,41 +3836,46 @@ "open_weights": true, "limit": { "context": 131072, - "output": 8192 + "output": 16384 }, "cost": { - "input": 0.175, - "output": 0.7 + "input": 0.287, + "output": 1.147, + "reasoning": 2.868 } }, - "qwen2-5-14b-instruct": { - "id": "qwen2-5-14b-instruct", - "name": "Qwen2.5 14B Instruct", + "qwen-omni-turbo-realtime": { + "id": "qwen-omni-turbo-realtime", + "name": "Qwen-Omni Turbo Realtime", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 32768, + "output": 2048 }, "cost": { - "input": 0.35, - "output": 1.4 + "input": 0.23, + "output": 0.918, + "input_audio": 3.584, + "output_audio": 7.168 } }, - "qwen2-5-32b-instruct": { - "id": "qwen2-5-32b-instruct", - "name": "Qwen2.5 32B Instruct", + "qwen2-5-math-7b-instruct": { + "id": "qwen2-5-math-7b-instruct", + "name": "Qwen2.5-Math 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, @@ -3306,17 +3890,18 @@ }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 4096, + "output": 3072 }, "cost": { - "input": 0.7, - "output": 2.8 + "input": 0.144, + "output": 0.287 } }, "qwen3-next-80b-a3b-instruct": { "id": "qwen3-next-80b-a3b-instruct", "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, @@ -3335,325 +3920,380 @@ "output": 32768 }, "cost": { - "input": 0.5, - "output": 2 + "input": 0.144, + "output": 0.574 } }, - "qwen-plus-character-ja": { - "id": "qwen-plus-character-ja", - "name": "Qwen Plus Character (Japanese)", + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01", - "last_updated": "2024-01", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 512 + "context": 1000000, + "output": 64000 }, "cost": { "input": 0.5, - "output": 1.4 + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "qwen3-omni-flash-realtime": { - "id": "qwen3-omni-flash-realtime", - "name": "Qwen3-Omni Flash Realtime", + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.52, - "output": 1.99, - "input_audio": 4.57, - "output_audio": 18.13 + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 } }, - "qwen3-vl-30b-a3b": { - "id": "qwen3-vl-30b-a3b", - "name": "Qwen3-VL 30B-A3B", + "qwen-long": { + "id": "qwen-long", + "name": "Qwen Long", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-04", + "release_date": "2025-01-25", + "last_updated": "2025-01-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 10000000, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.8, - "reasoning": 2.4 + "input": 0.072, + "output": 0.287 } }, - "qwen3-vl-plus": { - "id": "qwen3-vl-plus", - "name": "Qwen3-VL Plus", + "qwen-math-turbo": { + "id": "qwen-math-turbo", + "name": "Qwen Math Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "knowledge": "2024-04", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 4096, + "output": 3072 }, "cost": { - "input": 0.2, - "output": 1.6, - "reasoning": 4.8 + "input": 0.287, + "output": 0.861 } }, - "qwen3-coder-480b-a35b-instruct": { - "id": "qwen3-coder-480b-a35b-instruct", - "name": "Qwen3-Coder 480B-A35B Instruct", + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, "output": 65536 }, "cost": { - "input": 1.5, - "output": 7.5 + "input": 0.861, + "output": 3.441 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3-Coder 30B-A3B Instruct", + "qwen2-5-omni-7b": { + "id": "qwen2-5-omni-7b", + "name": "Qwen2.5-Omni 7B", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-04", + "release_date": "2024-12", + "last_updated": "2024-12", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 32768, + "output": 2048 }, "cost": { - "input": 0.45, - "output": 2.25 + "input": 0.087, + "output": 0.345, + "input_audio": 5.448 } }, - "qwen-turbo": { - "id": "qwen-turbo", - "name": "Qwen Turbo", + "qwen3-8b": { + "id": "qwen3-8b", + "name": "Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 38912 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-11-01", - "last_updated": "2025-04-28", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.2, - "reasoning": 0.5 + "input": 0.072, + "output": 0.287, + "reasoning": 0.717 } }, - "qwen-mt-turbo": { - "id": "qwen-mt-turbo", - "name": "Qwen-MT Turbo", + "qwen2-5-14b-instruct": { + "id": "qwen2-5-14b-instruct", + "name": "Qwen2.5 14B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2025-01", - "last_updated": "2025-01", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, + "context": 131072, "output": 8192 }, "cost": { - "input": 0.16, - "output": 0.49 + "input": 0.144, + "output": 0.431 } }, - "qwen3.6-max-preview": { - "id": "qwen3.6-max-preview", - "name": "Qwen3.6 Max Preview", - "family": "qwen", + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 131072 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "release_date": "2026-04-14", + "last_updated": "2026-04-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, - "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 1.3, - "output": 7.8, - "cache_read": 0.13, - "cache_write": 1.625 - } - }, - "qwen2-5-omni-7b": { - "id": "qwen2-5-omni-7b", - "name": "Qwen2.5-Omni 7B", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-12", - "last_updated": "2024-12", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] - }, "open_weights": true, "limit": { - "context": 32768, - "output": 2048 + "context": 202752, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4, - "input_audio": 6.76 + "input": 0.87, + "output": 3.48, + "cache_read": 0.17 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", - "family": "qwen", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 1000000, - "output": 65536 + "output": 384000 }, "cost": { - "input": 0.4, - "output": 2.4, - "reasoning": 2.4 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "qwen2-5-vl-72b-instruct": { - "id": "qwen2-5-vl-72b-instruct", - "name": "Qwen2.5-VL 72B Instruct", + "qwen3-next-80b-a3b-thinking": { + "id": "qwen3-next-80b-a3b-thinking", + "name": "Qwen3-Next 80B-A3B (Thinking)", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 131072, - "output": 8192 + "output": 32768 }, "cost": { - "input": 2.8, - "output": 8.4 + "input": 0.144, + "output": 1.434 } }, "qvq-max": { "id": "qvq-max", "name": "QVQ Max", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "family": "qvq", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "knowledge": "2024-04", @@ -3669,697 +4309,805 @@ "output": 8192 }, "cost": { - "input": 1.2, - "output": 4.8 + "input": 1.147, + "output": 4.588 } }, - "qwen3-14b": { - "id": "qwen3-14b", - "name": "Qwen3 14B", + "qwen-plus-character": { + "id": "qwen-plus-character", + "name": "Qwen Plus Character", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-04", + "release_date": "2024-01", + "last_updated": "2024-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 32768, + "output": 4096 }, "cost": { - "input": 0.35, - "output": 1.4, - "reasoning": 4.2 + "input": 0.115, + "output": 0.287 } }, - "qwen3.5-35b-a3b": { - "id": "qwen3.5-35b-a3b", - "name": "Qwen3.5 35B-A3B", - "family": "qwen", - "attachment": true, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Moonshot Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2026-02-23", - "last_updated": "2026-02-23", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 65536 + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2 + "input": 0.574, + "output": 2.294 } - } - } - }, - "scaleway": { - "id": "scaleway", - "env": ["SCALEWAY_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.scaleway.ai/v1", - "name": "Scaleway", - "doc": "https://www.scaleway.com/en/docs/generative-apis/", - "models": { - "qwen3-embedding-8b": { - "id": "qwen3-embedding-8b", - "name": "Qwen3 Embedding 8B", - "family": "qwen", + }, + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-25-11", - "last_updated": "2026-03-17", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 4096 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.574, + "output": 2.294 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", + "qwen3.5-flash": { + "id": "qwen3.5-flash", + "name": "Qwen3.5 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-07-01", - "last_updated": "2026-03-17", + "knowledge": "2025-04", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 260000, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.75, - "output": 2.25 + "input": 0.172, + "output": 1.72, + "reasoning": 1.72 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": true, - "reasoning": false, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2026-03-17", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 100000, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.9, - "output": 0.9 + "input": 1.1, + "output": 3.851, + "cache_read": 0.275, + "cache_write": 0 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", - "family": "qwen", - "attachment": false, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 131072 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.1875, + "output": 1.125, + "cache_write": 0.234375 } }, - "devstral-2-123b-instruct-2512": { - "id": "devstral-2-123b-instruct-2512", - "name": "Devstral 2 123B Instruct (2512)", - "family": "devstral", + "qwen2-5-vl-72b-instruct": { + "id": "qwen2-5-vl-72b-instruct", + "name": "Qwen2.5-VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-01-07", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 2 + "input": 2.294, + "output": 6.881 } }, - "deepseek-r1-distill-llama-70b": { - "id": "deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", + "qwen3-vl-plus": { + "id": "qwen3-vl-plus", + "name": "Qwen3-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2026-03-17", + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 8196 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.9, - "output": 0.9 + "input": 0.143353, + "output": 1.433525, + "reasoning": 4.300576 } }, - "pixtral-12b-2409": { - "id": "pixtral-12b-2409", - "name": "Pixtral 12B 2409", - "family": "pixtral", - "attachment": true, + "qwen-vl-ocr": { + "id": "qwen-vl-ocr", + "name": "Qwen-VL OCR", + "description": "OCR model for extracting structured text from documents and screenshots", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2024-09-25", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2024-10-28", + "last_updated": "2025-04-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 34096, "output": 4096 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.717, + "output": 0.717 } }, - "whisper-large-v3": { - "id": "whisper-large-v3", - "name": "Whisper Large v3", - "family": "whisper", + "qwen-mt-turbo": { + "id": "qwen-mt-turbo", + "name": "Qwen-MT Turbo", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2023-09-01", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-01", + "last_updated": "2025-01", "modalities": { - "input": ["audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, + "context": 16384, "output": 8192 }, "cost": { - "input": 0.003, - "output": 0 + "input": 0.101, + "output": 0.28 } }, - "voxtral-small-24b-2507": { - "id": "voxtral-small-24b-2507", - "name": "Voxtral Small 24B 2507", - "family": "voxtral", - "attachment": true, + "qwen-math-plus": { + "id": "qwen-math-plus", + "name": "Qwen Math Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-07-01", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2024-08-16", + "last_updated": "2024-09-19", "modalities": { - "input": ["text", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 16384 + "context": 4096, + "output": 3072 }, "cost": { - "input": 0.15, - "output": 0.35 + "input": 0.574, + "output": 1.721 } }, - "gemma-3-27b-it": { - "id": "gemma-3-27b-it", - "name": "Gemma-3-27B-IT", - "family": "gemma", - "attachment": true, - "reasoning": true, - "tool_call": true, + "qwen-mt-plus": { + "id": "qwen-mt-plus", + "name": "Qwen-MT Plus", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2025-01", + "last_updated": "2025-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 40000, + "context": 16384, "output": 8192 }, "cost": { - "input": 0.25, - "output": 0.5 + "input": 0.259, + "output": 0.775 } }, - "bge-multilingual-gemma2": { - "id": "bge-multilingual-gemma2", - "name": "BGE Multilingual Gemma2", - "family": "gemma", + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-07-26", - "last_updated": "2025-06-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 3072 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.573, + "output": 3.44, + "reasoning": 3.44 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3-Coder 30B-A3B Instruct", + "qwen-omni-turbo": { + "id": "qwen-omni-turbo", + "name": "Qwen-Omni Turbo", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2025-01-19", + "last_updated": "2025-03-26", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 32768, + "output": 2048 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.058, + "output": 0.23, + "input_audio": 3.584, + "output_audio": 7.168 } }, - "mistral-small-3.2-24b-instruct-2506": { - "id": "mistral-small-3.2-24b-instruct-2506", - "name": "Mistral Small 3.2 24B Instruct (2506)", - "family": "mistral-small", + "qwen2-5-72b-instruct": { + "id": "qwen2-5-72b-instruct", + "name": "Qwen2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-06-20", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.35 + "input": 0.574, + "output": 1.721 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT-OSS 120B", - "family": "gpt-oss", - "attachment": true, - "reasoning": false, + "deepseek-r1-distill-qwen-7b": { + "id": "deepseek-r1-distill-qwen-7b", + "name": "DeepSeek R1 Distill Qwen 7B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2026-03-17", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.072, + "output": 0.144 } }, - "mistral-nemo-instruct-2407": { - "id": "mistral-nemo-instruct-2407", - "name": "Mistral Nemo Instruct 2407", - "family": "mistral-nemo", - "attachment": true, + "deepseek-v3-1": { + "id": "deepseek-v3-1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-07-25", - "last_updated": "2026-03-17", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.574, + "output": 1.721 } }, - "llama-3.1-8b-instruct": { - "id": "llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "qwen2-5-coder-32b-instruct": { + "id": "qwen2-5-coder-32b-instruct", + "name": "Qwen2.5-Coder 32B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2026-03-17", + "knowledge": "2024-04", + "release_date": "2024-11", + "last_updated": "2024-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.287, + "output": 0.861 } - } - } - }, - "nano-gpt": { - "id": "nano-gpt", - "env": ["NANO_GPT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://nano-gpt.com/api/v1", - "name": "NanoGPT", - "doc": "https://docs.nano-gpt.com", - "models": { - "glm-4-flash": { - "id": "glm-4-flash", - "name": "GLM-4 Flash", + }, + "qwen-flash": { + "id": "qwen-flash", + "name": "Qwen Flash", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-08-01", - "last_updated": "2024-08-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.1003, - "output": 0.1003 + "input": 0.022, + "output": 0.216 } }, - "Meta-Llama-3-1-8B-Instruct-FP8": { - "id": "Meta-Llama-3-1-8B-Instruct-FP8", - "name": "Llama 3.1 8B (decentralized)", + "deepseek-v3-2-exp": { + "id": "deepseek-v3-2-exp", + "name": "DeepSeek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.02, - "output": 0.03 + "input": 0.287, + "output": 0.431 } }, - "claude-opus-4-thinking:32000": { - "id": "claude-opus-4-thinking:32000", - "name": "Claude 4 Opus Thinking (32K)", - "attachment": true, - "reasoning": true, + "moonshot-kimi-k2-instruct": { + "id": "moonshot-kimi-k2-instruct", + "name": "Moonshot Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": false, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.574, + "output": 2.294 } }, - "gemini-2.5-pro-preview-05-06": { - "id": "gemini-2.5-pro-preview-05-06", - "name": "Gemini 2.5 Pro Preview 0506", - "attachment": true, + "deepseek-r1-distill-qwen-14b": { + "id": "deepseek-r1-distill-qwen-14b", + "name": "DeepSeek R1 Distill Qwen 14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-06", - "last_updated": "2025-05-06", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 32768, + "output": 16384 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.144, + "output": 0.431 } }, - "grok-3-mini-fast-beta": { - "id": "grok-3-mini-fast-beta", - "name": "Grok 3 Mini Fast Beta", + "qwen3-vl-235b-a22b": { + "id": "qwen3-vl-235b-a22b", + "name": "Qwen3-VL 235B-A22B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "input": 131072, - "output": 131072 + "output": 32768 }, "cost": { - "input": 0.6, - "output": 4 + "input": 0.286705, + "output": 1.14682, + "reasoning": 2.867051 } }, - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax M2", + "qwen-deep-research": { + "id": "qwen-deep-research", + "name": "Qwen Deep Research", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-10-25", - "last_updated": "2025-10-25", + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-01", + "last_updated": "2024-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 131072 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.17, - "output": 1.53 + "input": 7.742, + "output": 23.367 } }, - "command-a-reasoning-08-2025": { - "id": "command-a-reasoning-08-2025", - "name": "Cohere Command A (08/2025)", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Moonshot Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": false, - "release_date": "2025-08-22", - "last_updated": "2025-08-22", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.574, + "output": 2.411 } }, - "brave": { - "id": "brave", - "name": "Brave (Answers)", + "qwen3-vl-30b-a3b": { + "id": "qwen3-vl-30b-a3b", + "name": "Qwen3-VL 30B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2023-03-02", - "last_updated": "2024-01-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 5, - "output": 5 + "input": 0.108, + "output": 0.431, + "reasoning": 1.076 } }, - "exa-research": { - "id": "exa-research", - "name": "Exa (Research)", + "qwen-vl-max": { + "id": "qwen-vl-max", + "name": "Qwen-VL Max", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-04", - "last_updated": "2025-06-04", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-04-08", + "last_updated": "2025-08-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 8192, + "context": 131072, "output": 8192 }, "cost": { - "input": 2.5, - "output": 2.5 + "input": 0.23, + "output": 0.574 } }, - "Llama-3.3-70B-Nova": { - "id": "Llama-3.3-70B-Nova", - "name": "Llama 3.3 70B Nova", + "deepseek-r1-distill-qwen-1-5b": { + "id": "deepseek-r1-distill-qwen-1-5b", + "name": "DeepSeek R1 Distill Qwen 1.5B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] @@ -4367,887 +5115,1072 @@ "open_weights": false, "limit": { "context": 32768, - "input": 32768, "output": 16384 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0, + "output": 0 } }, - "gemini-exp-1206": { - "id": "gemini-exp-1206", - "name": "Gemini 2.0 Pro 1206", - "attachment": true, + "qwen-max": { + "id": "qwen-max", + "name": "Qwen Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-04-03", + "last_updated": "2025-01-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2097152, - "input": 2097152, + "context": 131072, "output": 8192 }, "cost": { - "input": 1.258, - "output": 4.998 + "input": 0.345, + "output": 1.377 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "Claude 4.5 Opus", - "attachment": true, + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 0.3, + "output": 1.2 } }, - "auto-model-basic": { - "id": "auto-model-basic", - "name": "Auto model (Basic)", + "qwen3-235b-a22b": { + "id": "qwen3-235b-a22b", + "name": "Qwen3 235B-A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 38912 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 1000000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.287, + "output": 1.147, + "reasoning": 2.868 } }, - "jamba-mini": { - "id": "jamba-mini", - "name": "Jamba Mini", + "deepseek-r1-distill-llama-8b": { + "id": "deepseek-r1-distill-llama-8b", + "name": "DeepSeek R1 Distill Llama 8B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.1989, - "output": 0.408 + "input": 0, + "output": 0 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview (09/2025)", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Moonshot Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.1, - "output": 0.4 - } - }, - "yi-large": { - "id": "yi-large", - "name": "Yi Large", - "attachment": false, - "reasoning": false, - "tool_call": false, "structured_output": false, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, - "output": 4096 + "context": 262144, + "output": 16384 }, "cost": { - "input": 3.196, - "output": 3.196 + "input": 0.929, + "output": 3.858 } }, - "auto-model-premium": { - "id": "auto-model-premium", - "name": "Auto model (Premium)", + "qwen3-coder-480b-a35b-instruct": { + "id": "qwen3-coder-480b-a35b-instruct", + "name": "Qwen3-Coder 480B-A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 1000000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.861, + "output": 3.441 } }, - "azure-gpt-4o": { - "id": "azure-gpt-4o", - "name": "Azure gpt-4o", - "attachment": true, - "reasoning": false, + "qwq-plus": { + "id": "qwq-plus", + "name": "QwQ Plus", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 2.499, - "output": 9.996 + "input": 0.23, + "output": 0.574 } }, - "deepseek-v3-0324": { - "id": "deepseek-v3-0324", - "name": "DeepSeek Chat 0324", + "qwen2-5-32b-instruct": { + "id": "qwen2-5-32b-instruct", + "name": "Qwen2.5 32B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, + "context": 131072, "output": 8192 }, "cost": { - "input": 0.25, - "output": 0.7 + "input": 0.287, + "output": 0.861 } }, - "claude-3-5-haiku-20241022": { - "id": "claude-3-5-haiku-20241022", - "name": "Claude 3.5 Haiku", - "attachment": true, - "reasoning": false, + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 4 + "input": 0.43, + "output": 2.58, + "reasoning": 2.58 } }, - "doubao-seed-1-8-251215": { - "id": "doubao-seed-1-8-251215", - "name": "Doubao Seed 1.8", + "tongyi-intent-detect-v3": { + "id": "tongyi-intent-detect-v3", + "name": "Tongyi Intent Detect V3", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "yi", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-15", - "last_updated": "2025-12-15", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-01", + "last_updated": "2024-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 8192, + "output": 1024 }, "cost": { - "input": 0.612, - "output": 6.12 + "input": 0.058, + "output": 0.144 } }, - "doubao-seed-1-6-250615": { - "id": "doubao-seed-1-6-250615", - "name": "Doubao Seed 1.6", + "qwen3-coder-flash": { + "id": "qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-15", - "last_updated": "2025-06-15", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.204, - "output": 0.51 + "input": 0.144, + "output": 0.574 } }, - "ernie-x1.1-preview": { - "id": "ernie-x1.1-preview", - "name": "ERNIE X1.1", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-10", - "last_updated": "2025-09-10", + "deepseek-r1-distill-qwen-32b": { + "id": "deepseek-r1-distill-qwen-32b", + "name": "DeepSeek R1 Distill Qwen 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 64000, - "input": 64000, - "output": 8192 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.287, + "output": 0.861 } }, - "ernie-5.0-thinking-preview": { - "id": "ernie-5.0-thinking-preview", - "name": "Ernie 5.0 Thinking Preview", - "attachment": true, + "qwq-32b": { + "id": "qwq-32b", + "name": "QwQ 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-12", + "last_updated": "2024-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 1.1, - "output": 2 + "input": 0.287, + "output": 0.861 } }, - "glm-4-air-0111": { - "id": "glm-4-air-0111", - "name": "GLM 4 Air 0111", + "qwen3-14b": { + "id": "qwen3-14b", + "name": "Qwen3 14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-11", - "last_updated": "2025-01-11", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 38912 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.1394, - "output": 0.1394 + "input": 0.144, + "output": 0.574, + "reasoning": 1.434 } }, - "fastgpt": { - "id": "fastgpt", - "name": "Web Answer", + "qwen3-asr-flash": { + "id": "qwen3-asr-flash", + "name": "Qwen3-ASR Flash", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2023-08-01", - "last_updated": "2024-01-01", + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-09-08", + "last_updated": "2025-09-08", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 53248, + "output": 4096 }, "cost": { - "input": 7.5, - "output": 7.5 + "input": 0.032, + "output": 0.032 } }, - "doubao-seed-1-6-thinking-250615": { - "id": "doubao-seed-1-6-thinking-250615", - "name": "Doubao Seed 1.6 Thinking", + "qwen-doc-turbo": { + "id": "qwen-doc-turbo", + "name": "Qwen Doc Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-15", - "last_updated": "2025-06-15", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-01", + "last_updated": "2024-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.204, - "output": 2.04 + "input": 0.087, + "output": 0.144 } }, - "gemini-2.0-flash-001": { - "id": "gemini-2.0-flash-001", - "name": "Gemini 2.0 Flash", - "attachment": true, - "reasoning": false, + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 32768 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 8192 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.1003, - "output": 0.408 + "input": 0.86, + "output": 3.15 } }, - "claude-opus-4-1-thinking:32000": { - "id": "claude-opus-4-1-thinking:32000", - "name": "Claude 4.1 Opus Thinking (32K)", - "attachment": true, + "qwen-turbo": { + "id": "qwen-turbo", + "name": "Qwen Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 38912 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-11-01", + "last_updated": "2025-07-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.044, + "output": 0.087, + "reasoning": 0.431 } }, - "Llama-3.3-70B-RAWMAW": { - "id": "Llama-3.3-70B-RAWMAW", - "name": "Llama 3.3 70B RAWMAW", + "qwen2-5-7b-instruct": { + "id": "qwen2-5-7b-instruct", + "name": "Qwen2.5 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.072, + "output": 0.144 } }, - "GLM-4.5-Air-Derestricted-Steam": { - "id": "GLM-4.5-Air-Derestricted-Steam", - "name": "GLM 4.5 Air Derestricted Steam", + "qwen2-5-vl-7b-instruct": { + "id": "qwen2-5-vl-7b-instruct", + "name": "Qwen2.5-VL 7B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 220600, - "input": 220600, - "output": 65536 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.287, + "output": 0.717 } }, - "claude-3-5-sonnet-20241022": { - "id": "claude-3-5-sonnet-20241022", - "name": "Claude 3.5 Sonnet", - "attachment": true, - "reasoning": false, + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 131072 + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 8192 + "context": 245800, + "output": 65536 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 1.32, + "output": 7.9, + "cache_read": 0.132 } }, - "yi-medium-200k": { - "id": "yi-medium-200k", - "name": "Yi Medium 200k", + "qwen-vl-plus": { + "id": "qwen-vl-plus", + "name": "Qwen-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-03-01", - "last_updated": "2024-03-01", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-08-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 2.499, - "output": 2.499 + "input": 0.115, + "output": 0.287 } }, - "Gemma-3-27B-ArliAI-RPMax-v3": { - "id": "Gemma-3-27B-ArliAI-RPMax-v3", - "name": "Gemma 3 27B RPMax v3", + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-03", - "last_updated": "2025-07-03", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } } }, - "phi-4-mini-instruct": { - "id": "phi-4-mini-instruct", - "name": "Phi 4 Mini", + "MiniMax/MiniMax-M2.7": { + "id": "MiniMax/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "Llama-3.3+(3v3.3)-70B-TenyxChat-DaybreakStorywriter": { - "id": "Llama-3.3+(3v3.3)-70B-TenyxChat-DaybreakStorywriter", - "name": "Llama 3.3+ 70B TenyxChat DaybreakStorywriter", + "kimi/kimi-k2.5": { + "id": "kimi/kimi-k2.5", + "name": "kimi/kimi-k2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "ernie-x1-32k": { - "id": "ernie-x1-32k", - "name": "Ernie X1 32k", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "siliconflow/deepseek-r1-0528": { + "id": "siliconflow/deepseek-r1-0528", + "name": "siliconflow/deepseek-r1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-05-28", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 16384 + "context": 163840, + "output": 32768 }, "cost": { - "input": 0.33, - "output": 1.32 + "input": 0.5, + "output": 2.18 } }, - "deepseek-chat": { - "id": "deepseek-chat", - "name": "DeepSeek V3/Deepseek Chat", - "attachment": true, - "reasoning": false, + "siliconflow/deepseek-v3.1-terminus": { + "id": "siliconflow/deepseek-v3.1-terminus", + "name": "siliconflow/deepseek-v3.1-terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "temperature": true, + "release_date": "2025-09-29", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 163840, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 0.7 + "input": 0.27, + "output": 1 } }, - "glm-z1-air": { - "id": "glm-z1-air", - "name": "GLM Z1 Air", + "siliconflow/deepseek-v3-0324": { + "id": "siliconflow/deepseek-v3-0324", + "name": "siliconflow/deepseek-v3-0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "temperature": true, + "release_date": "2024-12-26", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 16384 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.07, - "output": 0.07 + "input": 0.25, + "output": 1 } }, - "claude-3-7-sonnet-thinking:128000": { - "id": "claude-3-7-sonnet-thinking:128000", - "name": "Claude 3.7 Sonnet Thinking (128K)", - "attachment": true, + "siliconflow/deepseek-v3.2": { + "id": "siliconflow/deepseek-v3.2", + "name": "siliconflow/deepseek-v3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-12-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 163840, + "output": 65536 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.27, + "output": 0.42 } }, - "glm-4-air": { - "id": "glm-4-air", - "name": "GLM-4 Air", + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Hosted Qwen coder for software agents, repo edits, and long-context code", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-06-05", - "last_updated": "2024-06-05", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.2006, - "output": 0.2006 + "input": 1, + "output": 5 } - }, - "Llama-3.3-70B-MiraiFanfare": { - "id": "Llama-3.3-70B-MiraiFanfare", - "name": "Llama 3.3 70b Mirai Fanfare", + } + } + }, + "regolo-ai": { + "id": "regolo-ai", + "env": ["REGOLO_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.regolo.ai/v1", + "name": "Regolo AI", + "doc": "https://docs.regolo.ai/", + "models": { + "llama-3.1-8b-instruct": { + "id": "llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "temperature": true, + "release_date": "2025-04-07", + "last_updated": "2025-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 120000, + "output": 120000 }, "cost": { - "input": 0.493, - "output": 0.493 + "input": 0.05, + "output": 0.25 } }, - "gemini-2.0-flash-thinking-exp-01-21": { - "id": "gemini-2.0-flash-thinking-exp-01-21", - "name": "Gemini 2.0 Flash Thinking 0121", - "attachment": true, + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax 2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-21", - "last_updated": "2025-01-21", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-10", + "last_updated": "2026-03-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 8192 + "context": 190000, + "output": 64000 }, "cost": { - "input": 0.306, - "output": 1.003 + "input": 0.8, + "output": 3.5 } }, - "Magistral-Small-2506": { - "id": "Magistral-Small-2506", - "name": "Magistral Small 2506", + "mistral-small3.2": { + "id": "mistral-small3.2", + "name": "Mistral Small 3.2", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 120000, + "output": 120000 }, "cost": { - "input": 0.4, - "output": 1.4 + "input": 0.5, + "output": 2.2 } }, - "doubao-1.5-pro-32k": { - "id": "doubao-1.5-pro-32k", - "name": "Doubao 1.5 Pro 32k", + "qwen3-reranker-4b": { + "id": "qwen3-reranker-4b", + "name": "Qwen3-Reranker-4B", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-01-22", - "last_updated": "2025-01-22", + "temperature": false, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, + "context": 32768, "output": 8192 }, "cost": { - "input": 0.1343, - "output": 0.3349 + "input": 0.12, + "output": 0.12 } }, - "venice-uncensored:web": { - "id": "venice-uncensored:web", - "name": "Venice Uncensored Web", + "qwen3-embedding-8b": { + "id": "qwen3-embedding-8b", + "name": "Qwen3-Embedding-8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-05-01", - "last_updated": "2024-05-01", + "temperature": false, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 80000, - "input": 80000, - "output": 16384 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 0.1, + "output": 0.1 } }, - "glm-4": { - "id": "glm-4", - "name": "GLM-4", + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-01-16", - "last_updated": "2024-01-16", + "tool_call": true, + "temperature": true, + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] @@ -5255,479 +6188,531 @@ "open_weights": false, "limit": { "context": 128000, - "input": 128000, - "output": 4096 + "output": 16384 }, "cost": { - "input": 14.994, - "output": 14.994 + "input": 0.6, + "output": 2.7 } }, - "qwen-max": { - "id": "qwen-max", - "name": "Qwen 2.5 Max", + "qwen-image": { + "id": "qwen-image", + "name": "Qwen-Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-04-03", - "last_updated": "2024-04-03", + "temperature": true, + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 8192 + "context": 8192, + "output": 4096 }, "cost": { - "input": 1.5997, - "output": 6.392 + "input": 0.5, + "output": 2 } }, - "qwen3-vl-235b-a22b-instruct-original": { - "id": "qwen3-vl-235b-a22b-instruct-original", - "name": "Qwen3 VL 235B A22B Instruct Original", + "qwen3.5-122b": { + "id": "qwen3.5-122b", + "name": "Qwen3.5-122B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 1.2 + "input": 0.9, + "output": 3.6 } }, - "jamba-large-1.6": { - "id": "jamba-large-1.6", - "name": "Jamba Large 1.6", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT-OSS-120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.989, - "output": 7.99 + "input": 1, + "output": 4.2 } }, - "qwen-plus": { - "id": "qwen-plus", - "name": "Qwen Plus", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "Qwen3-Coder-Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 995904, - "input": 995904, - "output": 32768 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.3995, - "output": 1.2002 + "input": 0.3, + "output": 1.2 } }, - "qwen25-vl-72b-instruct": { - "id": "qwen25-vl-72b-instruct", - "name": "Qwen25 VL 72b", + "qwen3.5-9b": { + "id": "qwen3.5-9b", + "name": "Qwen3.5-9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-10", - "last_updated": "2025-05-10", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, - "output": 32768 + "context": 262144, + "output": 8192 }, "cost": { - "input": 0.69989, - "output": 0.69989 + "input": 0.15, + "output": 0.6 } }, - "claude-sonnet-4-thinking:64000": { - "id": "claude-sonnet-4-thinking:64000", - "name": "Claude 4 Sonnet Thinking (64K)", + "mistral-small-4-119b": { + "id": "mistral-small-4-119b", + "name": "Mistral Small 4 119B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 + "context": 256000, + "output": 16384 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.75, + "output": 3 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro", - "attachment": true, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "GPT-OSS-20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "temperature": true, + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 12 + "input": 0.4, + "output": 1.8 } - }, - "Llama-3.3+(3.1v3.3)-70B-New-Dawn-v1.1": { - "id": "Llama-3.3+(3.1v3.3)-70B-New-Dawn-v1.1", - "name": "Llama 3.3+ 70B New Dawn v1.1", + } + } + }, + "stackit": { + "id": "stackit", + "env": ["STACKIT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.openai-compat.model-serving.eu01.onstackit.cloud/v1", + "name": "STACKIT", + "doc": "https://docs.stackit.cloud/products/data-and-ai/ai-model-serving/basics/available-shared-models", + "models": { + "cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic": { + "id": "cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic", + "name": "Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.53, + "output": 0.76 } }, - "GLM-4.5-Air-Derestricted-Iceblink-ReExtract": { - "id": "GLM-4.5-Air-Derestricted-Iceblink-ReExtract", - "name": "GLM 4.5 Air Derestricted Iceblink ReExtract", - "attachment": false, + "google/gemma-3-27b-it": { + "id": "google/gemma-3-27b-it", + "name": "Gemma 3 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, "tool_call": false, "structured_output": false, - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "temperature": true, + "release_date": "2025-05-17", + "last_updated": "2025-05-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 98304 + "context": 37000, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.53, + "output": 0.76 } }, - "universal-summarizer": { - "id": "universal-summarizer", - "name": "Universal Summarizer", + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": false, - "release_date": "2023-05-01", - "last_updated": "2024-01-01", + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 262144, + "output": 16384 }, "cost": { - "input": 30, - "output": 30 + "input": 0.53, + "output": 0.76 } }, - "claude-sonnet-4-thinking:32768": { - "id": "claude-sonnet-4-thinking:32768", - "name": "Claude 4 Sonnet Thinking (32K)", + "Qwen/Qwen3-VL-Embedding-8B": { + "id": "Qwen/Qwen3-VL-Embedding-8B", + "name": "Qwen3-VL Embedding 8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": false, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 - }, + "context": 32000, + "output": 4096 + }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.09, + "output": 0.09 } }, - "sarvan-medium": { - "id": "sarvan-medium", - "name": "Sarvam Medium", - "attachment": false, + "Qwen/Qwen3-VL-235B-A22B-Instruct-FP8": { + "id": "Qwen/Qwen3-VL-235B-A22B-Instruct-FP8", + "name": "Qwen3-VL 235B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": false, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, + "context": 218000, "output": 16384 }, "cost": { - "input": 0.25, - "output": 0.75 + "input": 1.76, + "output": 2.05 } }, - "claude-3-7-sonnet-thinking:8192": { - "id": "claude-3-7-sonnet-thinking:8192", - "name": "Claude 3.7 Sonnet Thinking (8K)", - "attachment": true, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "input": 200000, - "output": 64000 - }, - "cost": { - "input": 2.992, - "output": 14.994 - } - }, - "gemini-2.5-flash-preview-05-20": { - "id": "gemini-2.5-flash-preview-05-20", - "name": "Gemini 2.5 Flash 0520", - "attachment": true, - "reasoning": false, - "tool_call": false, "structured_output": false, - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048000, - "input": 1048000, - "output": 65536 + "context": 131000, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.53, + "output": 0.76 } }, - "GLM-4.5-Air-Derestricted-Iceblink-v2-ReExtract": { - "id": "GLM-4.5-Air-Derestricted-Iceblink-v2-ReExtract", - "name": "GLM 4.5 Air Derestricted Iceblink v2 ReExtract", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": false, - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "input": 131072, - "output": 65536 + "output": 8192 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.18, + "output": 0.29 } }, - "Llama-3.3-70B-Fallen-v1": { - "id": "Llama-3.3-70B-Fallen-v1", - "name": "Llama 3.3 70B Fallen v1", + "intfloat/e5-mistral-7b-instruct": { + "id": "intfloat/e5-mistral-7b-instruct", + "name": "E5 Mistral 7B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "mistral", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": false, + "release_date": "2023-12-11", + "last_updated": "2023-12-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.02, + "output": 0.02 } - }, - "qwen3-vl-235b-a22b-thinking": { - "id": "qwen3-vl-235b-a22b-thinking", - "name": "Qwen3 VL 235B A22B Thinking", - "attachment": true, - "reasoning": true, + } + } + }, + "vercel": { + "id": "vercel", + "env": ["AI_GATEWAY_API_KEY"], + "npm": "@ai-sdk/gateway", + "name": "Vercel AI Gateway", + "doc": "https://github.com/vercel/ai/tree/5eb85cc45a259553501f535b8ac79a77d0e79223/packages/gateway", + "models": { + "xai/grok-imagine-video-1.5": { + "id": "xai/grok-imagine-video-1.5", + "name": "Grok Imagine Video 1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "temperature": true, + "release_date": "2026-06-22", + "last_updated": "2026-06-22", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 - }, - "cost": { - "input": 0.5, - "output": 6 + "context": 0, + "output": 0 } }, - "claude-3-7-sonnet-thinking:32768": { - "id": "claude-3-7-sonnet-thinking:32768", - "name": "Claude 3.7 Sonnet Thinking (32K)", - "attachment": true, + "xai/grok-4.1-fast-reasoning": { + "id": "xai/grok-4.1-fast-reasoning", + "name": "Grok 4.1 Fast Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-07-15", - "last_updated": "2025-07-15", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-19", + "last_updated": "2025-07-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 1000000, + "output": 1000000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "claude-3-7-sonnet-thinking:1024": { - "id": "claude-3-7-sonnet-thinking:1024", - "name": "Claude 3.7 Sonnet Thinking (1K)", + "xai/grok-4.20-non-reasoning-beta": { + "id": "xai/grok-4.20-non-reasoning-beta", + "name": "Grok 4.20 Beta Non-Reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 1.25, + "output": 2.5, + "cache_read": 0.4 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Claude Sonnet 4.5", + "xai/grok-4.3": { + "id": "xai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -5735,47 +6720,47 @@ "open_weights": false, "limit": { "context": 1000000, - "input": 1000000, - "output": 64000 + "output": 1000000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "Llama-3.3-70B-Vulpecula-R1": { - "id": "Llama-3.3-70B-Vulpecula-R1", - "name": "Llama 3.3 70B Vulpecula R1", + "xai/grok-tts": { + "id": "xai/grok-tts", + "name": "Grok TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "grok", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "claude-sonnet-4-thinking:8192": { - "id": "claude-sonnet-4-thinking:8192", - "name": "Claude 4 Sonnet Thinking (8K)", - "attachment": true, - "reasoning": true, + "xai/grok-4.1-fast-non-reasoning": { + "id": "xai/grok-4.1-fast-non-reasoning", + "name": "Grok 4.1 Fast Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-19", + "last_updated": "2025-07-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -5783,1175 +6768,1164 @@ "open_weights": false, "limit": { "context": 1000000, - "input": 1000000, - "output": 64000 + "output": 1000000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "attachment": true, - "reasoning": true, + "xai/grok-voice-think-fast-1.0": { + "id": "xai/grok-voice-think-fast-1.0", + "name": "Grok Voice Think Fast 1.0", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "grok", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "temperature": true, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 2.5, - "output": 10 + "context": 0, + "output": 0 } }, - "Llama-3.3-70B-Ignition-v0.1": { - "id": "Llama-3.3-70B-Ignition-v0.1", - "name": "Llama 3.3 70B Ignition v0.1", + "xai/grok-imagine-video": { + "id": "xai/grok-imagine-video", + "name": "Grok Imagine", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-01-28", + "last_updated": "2026-01-28", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "glm-4-plus-0111": { - "id": "glm-4-plus-0111", - "name": "GLM 4 Plus 0111", + "xai/grok-4.20-multi-agent-beta": { + "id": "xai/grok-4.20-multi-agent-beta", + "name": "Grok 4.20 Multi Agent Beta", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 9.996, - "output": 9.996 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "KAT-Coder-Air-V1": { - "id": "KAT-Coder-Air-V1", - "name": "KAT Coder Air V1", + "xai/grok-stt": { + "id": "xai/grok-stt", + "name": "Grok STT", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "grok", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 - }, - "cost": { - "input": 0.1, - "output": 0.2 + "context": 0, + "output": 0 } }, - "deepseek-r1-sambanova": { - "id": "deepseek-r1-sambanova", - "name": "DeepSeek R1 Fast", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-20", - "last_updated": "2025-02-20", + "xai/grok-4.5": { + "id": "xai/grok-4.5", + "name": "Grok 4.5", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 500000, + "output": 500000 }, "cost": { - "input": 4.998, - "output": 6.987 + "input": 2, + "output": 6, + "cache_read": 0.5 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek R1", - "attachment": false, + "xai/grok-4.20-reasoning": { + "id": "xai/grok-4.20-reasoning", + "name": "Grok 4.20 Reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-10", + "last_updated": "2026-03-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.4, - "output": 1.7 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "doubao-1-5-thinking-pro-250415": { - "id": "doubao-1-5-thinking-pro-250415", - "name": "Doubao 1.5 Thinking Pro", + "xai/grok-4.20-reasoning-beta": { + "id": "xai/grok-4.20-reasoning-beta", + "name": "Grok 4.20 Beta Reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-17", - "last_updated": "2025-04-17", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.6, - "output": 2.4 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "sonar-pro": { - "id": "sonar-pro", - "name": "Perplexity Pro", + "xai/grok-imagine-video-1.5-preview": { + "id": "xai/grok-imagine-video-1.5-preview", + "name": "Grok Imagine Video 1.5 Preview", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "temperature": true, + "release_date": "2026-05-30", + "last_updated": "2026-05-30", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 128000 - }, - "cost": { - "input": 2.992, - "output": 14.994 + "context": 0, + "output": 0 } }, - "Gemma-3-27B-it-Abliterated": { - "id": "Gemma-3-27B-it-Abliterated", - "name": "Gemma 3 27B IT Abliterated", - "attachment": false, + "xai/grok-4.20-non-reasoning": { + "id": "xai/grok-4.20-non-reasoning", + "name": "Grok 4.20 Non-Reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-03", - "last_updated": "2025-07-03", + "tool_call": true, + "temperature": true, + "release_date": "2026-03-10", + "last_updated": "2026-03-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 96000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.42, - "output": 0.42 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "deepseek-chat-cheaper": { - "id": "deepseek-chat-cheaper", - "name": "DeepSeek V3/Chat Cheaper", - "attachment": true, - "reasoning": false, + "xai/grok-4.20-multi-agent": { + "id": "xai/grok-4.20-multi-agent", + "name": "Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "temperature": true, + "release_date": "2026-03-10", + "last_updated": "2026-03-23", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.25, - "output": 0.7 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "gemini-2.0-pro-exp-02-05": { - "id": "gemini-2.0-pro-exp-02-05", - "name": "Gemini 2.0 Pro 0205", - "attachment": true, + "xai/grok-imagine-image": { + "id": "xai/grok-imagine-image", + "name": "Grok Imagine Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-05", - "last_updated": "2025-02-05", + "temperature": true, + "release_date": "2026-01-28", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 2097152, - "input": 2097152, - "output": 8192 - }, - "cost": { - "input": 1.989, - "output": 7.956 + "context": 0, + "output": 0 } }, - "azure-gpt-4o-mini": { - "id": "azure-gpt-4o-mini", - "name": "Azure gpt-4o-mini", + "xai/grok-build-0.1": { + "id": "xai/grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "temperature": true, + "release_date": "2026-05-20", + "last_updated": "2026-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.1496, - "output": 0.595 + "input": 1, + "output": 2, + "cache_read": 0.2 } }, - "Llama-3.3-70B-MS-Nevoria": { - "id": "Llama-3.3-70B-MS-Nevoria", - "name": "Llama 3.3 70B MS Nevoria", + "moonshotai/kimi-k2": { + "id": "moonshotai/kimi-k2", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "release_date": "2025-07-11", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.57, + "output": 2.3 } }, - "claude-opus-4-thinking": { - "id": "claude-opus-4-thinking", - "name": "Claude 4 Opus Thinking", + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-07-15", - "last_updated": "2025-07-15", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 256000, + "output": 32768 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "Llama-3.3-70B-Sapphira-0.1": { - "id": "Llama-3.3-70B-Sapphira-0.1", - "name": "Llama 3.3 70B Sapphira 0.1", + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 216144, + "output": 216144 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.47, + "output": 2, + "cache_read": 0.141 } }, - "doubao-seed-code-preview-latest": { - "id": "doubao-seed-code-preview-latest", - "name": "Doubao Seed Code Preview", - "attachment": false, + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-26", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 262114, + "output": 262114 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "qwen-3.6-plus": { - "id": "qwen-3.6-plus", - "name": "Qwen 3.6 Plus", - "family": "qwen3.6", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 991800, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.45, - "output": 2.7 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "Llama-3.3-70B-ArliAI-RPMax-v1.4": { - "id": "Llama-3.3-70B-ArliAI-RPMax-v1.4", - "name": "Llama 3.3 70B RPMax v1.4", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "moonshotai/kimi-k2.7-code-highspeed": { + "id": "moonshotai/kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code High Speed", + "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-06-15", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.9, + "output": 8, + "cache_read": 0.38 } }, - "mistral-small-31-24b-instruct": { - "id": "mistral-small-31-24b-instruct", - "name": "Mistral Small 31 24b Instruct", - "attachment": true, + "klingai/kling-v3.0-motion-control": { + "id": "klingai/kling-v3.0-motion-control", + "name": "Kling v3.0 Motion Control", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ling", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "temperature": true, + "release_date": "2026-03-04", + "last_updated": "2026-03-04", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 131072 - }, - "cost": { - "input": 0.1, - "output": 0.3 + "context": 0, + "output": 0 } }, - "glm-4.1v-thinking-flashx": { - "id": "glm-4.1v-thinking-flashx", - "name": "GLM 4.1V Thinking FlashX", - "attachment": true, + "klingai/kling-v2.6-i2v": { + "id": "klingai/kling-v2.6-i2v", + "name": "Kling v2.6 Image-to-Video", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ling", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-12-21", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 64000, - "input": 64000, - "output": 8192 - }, - "cost": { - "input": 0.3, - "output": 0.3 + "context": 0, + "output": 0 } }, - "hunyuan-t1-latest": { - "id": "hunyuan-t1-latest", - "name": "Hunyuan T1", + "klingai/kling-v2.5-turbo-t2v": { + "id": "klingai/kling-v2.5-turbo-t2v", + "name": "Kling v2.5 Turbo Text-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ling", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-22", - "last_updated": "2025-03-22", + "temperature": true, + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 - }, - "cost": { - "input": 0.17, - "output": 0.66 + "context": 0, + "output": 0 } }, - "doubao-1-5-thinking-vision-pro-250428": { - "id": "doubao-1-5-thinking-vision-pro-250428", - "name": "Doubao 1.5 Thinking Vision Pro", - "attachment": true, + "klingai/kling-v3.0-i2v": { + "id": "klingai/kling-v3.0-i2v", + "name": "Kling v3.0 Image-to-Video", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ling", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-05-15", - "last_updated": "2025-05-15", + "temperature": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 0.55, - "output": 1.43 + "context": 0, + "output": 0 } }, - "asi1-mini": { - "id": "asi1-mini", - "name": "ASI1 Mini", - "attachment": true, + "klingai/kling-v2.5-turbo-i2v": { + "id": "klingai/kling-v2.5-turbo-i2v", + "name": "Kling v2.5 Turbo Image-to-Video", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ling", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "temperature": true, + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 1, - "output": 1 + "context": 0, + "output": 0 } }, - "ernie-5.0-thinking-latest": { - "id": "ernie-5.0-thinking-latest", - "name": "Ernie 5.0 Thinking", - "attachment": true, - "reasoning": true, + "klingai/kling-v3.0-t2v": { + "id": "klingai/kling-v3.0-t2v", + "name": "Kling v3.0 Text-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ling", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "temperature": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 1.1, - "output": 2 + "context": 0, + "output": 0 } }, - "Llama-3.3-70B-Incandescent-Malevolence": { - "id": "Llama-3.3-70B-Incandescent-Malevolence", - "name": "Llama 3.3 70B Incandescent Malevolence", + "klingai/kling-v2.6-motion-control": { + "id": "klingai/kling-v2.6-motion-control", + "name": "Kling v2.6 Motion Control", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ling", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-12-18", + "last_updated": "2025-12-21", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "Llama-3.3-70B-Damascus-R1": { - "id": "Llama-3.3-70B-Damascus-R1", - "name": "Damascus R1", + "klingai/kling-v2.6-t2v": { + "id": "klingai/kling-v2.6-t2v", + "name": "Kling v2.6 Text-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ling", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-12-21", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "Gemma-3-27B-Nidum-Uncensored": { - "id": "Gemma-3-27B-Nidum-Uncensored", - "name": "Gemma 3 27B Nidum Uncensored", + "voyage/voyage-4-lite": { + "id": "voyage/voyage-4-lite", + "name": "voyage-4-lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-03-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 96000 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 32000, + "output": 0 } }, - "gemini-2.5-flash-lite-preview-09-2025-thinking": { - "id": "gemini-2.5-flash-lite-preview-09-2025-thinking", - "name": "Gemini 2.5 Flash Lite Preview (09/2025) – Thinking", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "voyage/voyage-law-2": { + "id": "voyage/voyage-law-2", + "name": "voyage-law-2", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "voyage", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-04-15", + "last_updated": "2024-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 0.1, - "output": 0.4 + "context": 8192, + "output": 1536 } }, - "doubao-seed-2-0-pro-260215": { - "id": "doubao-seed-2-0-pro-260215", - "name": "Doubao Seed 2.0 Pro", + "voyage/voyage-4": { + "id": "voyage/voyage-4", + "name": "voyage-4", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-03-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 128000 - }, - "cost": { - "input": 0.782, - "output": 3.876 + "context": 32000, + "output": 0 } }, - "gemini-3-pro-image-preview": { - "id": "gemini-3-pro-image-preview", - "name": "Gemini 3 Pro Image", - "attachment": true, + "voyage/voyage-code-3": { + "id": "voyage/voyage-code-3", + "name": "voyage-code-3", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "voyage", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "temperature": true, + "release_date": "2024-12-04", + "last_updated": "2024-09", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 2, - "output": 12 + "context": 8192, + "output": 1536 } }, - "Gemma-3-27B-CardProjector-v4": { - "id": "Gemma-3-27B-CardProjector-v4", - "name": "Gemma 3 27B CardProjector v4", + "voyage/voyage-4-large": { + "id": "voyage/voyage-4-large", + "name": "voyage-4-large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-03-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 32000, + "output": 0 } }, - "jamba-mini-1.7": { - "id": "jamba-mini-1.7", - "name": "Jamba Mini 1.7", + "voyage/rerank-2.5": { + "id": "voyage/rerank-2.5", + "name": "Voyage Rerank 2.5", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "temperature": true, + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 - }, - "cost": { - "input": 0.1989, - "output": 0.408 + "context": 32000, + "output": 32000 } }, - "Llama-3.3-70B-Forgotten-Safeword-3.6": { - "id": "Llama-3.3-70B-Forgotten-Safeword-3.6", - "name": "Llama 3.3 70B Forgotten Safeword 3.6", + "voyage/rerank-2.5-lite": { + "id": "voyage/rerank-2.5-lite", + "name": "Voyage Rerank 2.5 Lite", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 32000, + "output": 32000 } }, - "doubao-1-5-thinking-pro-vision-250415": { - "id": "doubao-1-5-thinking-pro-vision-250415", - "name": "Doubao 1.5 Thinking Pro Vision", - "attachment": true, + "voyage/voyage-code-2": { + "id": "voyage/voyage-code-2", + "name": "voyage-code-2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "voyage", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 0.6, - "output": 2.4 - } - }, - "gemini-2.5-pro-preview-06-05": { - "id": "gemini-2.5-pro-preview-06-05", - "name": "Gemini 2.5 Pro Preview 0605", - "attachment": true, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "temperature": true, + "release_date": "2024-01-01", + "last_updated": "2024-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 2.5, - "output": 10 + "context": 8192, + "output": 1536 } }, - "gemini-2.0-pro-reasoner": { - "id": "gemini-2.0-pro-reasoner", - "name": "Gemini 2.0 Pro Reasoner", + "voyage/voyage-3.5-lite": { + "id": "voyage/voyage-3.5-lite", + "name": "voyage-3.5-lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-05", - "last_updated": "2025-02-05", + "temperature": true, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 - }, - "cost": { - "input": 1.292, - "output": 4.998 + "context": 8192, + "output": 1536 } }, - "doubao-seed-2-0-lite-260215": { - "id": "doubao-seed-2-0-lite-260215", - "name": "Doubao Seed 2.0 Lite", + "voyage/voyage-3.5": { + "id": "voyage/voyage-3.5", + "name": "voyage-3.5", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "temperature": true, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 32000 - }, - "cost": { - "input": 0.1462, - "output": 0.8738 + "context": 8192, + "output": 1536 } }, - "gemini-2.5-flash-lite-preview-06-17": { - "id": "gemini-2.5-flash-lite-preview-06-17", - "name": "Gemini 2.5 Flash Lite Preview", - "attachment": true, - "reasoning": true, + "voyage/voyage-3-large": { + "id": "voyage/voyage-3-large", + "name": "voyage-3-large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "voyage", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "temperature": true, + "release_date": "2025-01-07", + "last_updated": "2024-09", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 0.15, - "output": 0.6 + "context": 8192, + "output": 1536 } }, - "sonar-deep-research": { - "id": "sonar-deep-research", - "name": "Perplexity Deep Research", + "voyage/voyage-finance-2": { + "id": "voyage/voyage-finance-2", + "name": "voyage-finance-2", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "voyage", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-25", - "last_updated": "2025-02-25", + "temperature": true, + "release_date": "2024-06-03", + "last_updated": "2024-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 60000, - "input": 60000, - "output": 128000 - }, - "cost": { - "input": 3.4, - "output": 13.6 + "context": 8192, + "output": 1536 } }, - "Gemma-3-27B-it": { - "id": "Gemma-3-27B-it", - "name": "Gemma 3 27B IT", + "mistral/mistral-nemo": { + "id": "mistral/mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-07-18", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.15, + "output": 0.15 } }, - "Llama-3.3-70B-GeneticLemonade-Unleashed-v3": { - "id": "Llama-3.3-70B-GeneticLemonade-Unleashed-v3", - "name": "Llama 3.3 70B GeneticLemonade Unleashed v3", + "mistral/codestral-embed": { + "id": "mistral/codestral-embed", + "name": "Codestral Embed", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "codestral-embed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 8192, + "output": 1536 } }, - "Gemma-3-27B-Glitter": { - "id": "Gemma-3-27B-Glitter", - "name": "Gemma 3 27B Glitter", + "mistral/mistral-embed": { + "id": "mistral/mistral-embed", + "name": "Mistral Embed", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "mistral-embed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "temperature": true, + "release_date": "2023-12-11", + "last_updated": "2023-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 8192, + "output": 1536 } }, - "Llama-3.3-70B-The-Omega-Directive-Unslop-v2.1": { - "id": "Llama-3.3-70B-The-Omega-Directive-Unslop-v2.1", - "name": "Llama 3.3 70B Omega Directive Unslop v2.1", + "mistral/devstral-small": { + "id": "mistral/devstral-small", + "name": "Devstral Small 1.1", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-05-21", + "last_updated": "2025-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 64000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.1, + "output": 0.3 } }, - "qwen3-30b-a3b-instruct-2507": { - "id": "qwen3-30b-a3b-instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507", - "attachment": false, + "mistral/mistral-large-3": { + "id": "mistral/mistral-large-3", + "name": "Mistral Large 3", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-20", - "last_updated": "2025-02-20", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "input": 256000, - "output": 32768 + "output": 256000 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.5, + "output": 1.5 } }, - "gemini-2.5-flash-preview-09-2025-thinking": { - "id": "gemini-2.5-flash-preview-09-2025-thinking", - "name": "Gemini 2.5 Flash Preview (09/2025) – Thinking", - "attachment": true, + "mistral/mistral-medium-3.5": { + "id": "mistral/mistral-medium-3.5", + "name": "Mistral Medium Latest", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "release_date": "2026-04-29", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 1.5, + "output": 7.5 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", + "mistral/mistral-medium": { + "id": "mistral/mistral-medium", + "name": "Mistral Medium 3.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", "attachment": true, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 128000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 0.4, + "output": 2 } }, - "deepclaude": { - "id": "deepclaude", - "name": "DeepClaude", - "attachment": true, + "mistral/devstral-small-2": { + "id": "mistral/devstral-small-2", + "name": "Devstral Small 2", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-01", - "last_updated": "2025-02-01", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-09", + "last_updated": "2025-05-07", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 256000, + "output": 256000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.1, + "output": 0.3 } }, - "ernie-4.5-8k-preview": { - "id": "ernie-4.5-8k-preview", - "name": "Ernie 4.5 8k Preview", - "attachment": false, + "mistral/ministral-14b": { + "id": "mistral/ministral-14b", + "name": "Ministral 14B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-02", + "last_updated": "2025-12-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 16384 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.66, - "output": 2.6 + "input": 0.2, + "output": 0.2 } }, - "doubao-seed-2-0-mini-260215": { - "id": "doubao-seed-2-0-mini-260215", - "name": "Doubao Seed 2.0 Mini", + "mistral/devstral-2": { + "id": "mistral/devstral-2", + "name": "Devstral 2", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { "input": ["text"], "output": ["text"] @@ -6959,2215 +7933,2673 @@ "open_weights": false, "limit": { "context": 256000, - "input": 256000, - "output": 32000 + "output": 256000 }, "cost": { - "input": 0.0493, - "output": 0.4845 + "input": 0.4, + "output": 2 } }, - "gemini-3-pro-preview-thinking": { - "id": "gemini-3-pro-preview-thinking", - "name": "Gemini 3 Pro Thinking", + "mistral/mistral-small": { + "id": "mistral/mistral-small", + "name": "Mistral Small (latest)", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "temperature": true, + "knowledge": "2025-06", + "release_date": "2024-09-17", + "last_updated": "2026-03-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 32000, + "output": 4000 }, "cost": { - "input": 2, - "output": 12 + "input": 0.1, + "output": 0.3 } }, - "Llama-3.3-70B-GeneticLemonade-Opus": { - "id": "Llama-3.3-70B-GeneticLemonade-Opus", - "name": "Llama 3.3 70B GeneticLemonade Opus", + "mistral/ministral-8b": { + "id": "mistral/ministral-8b", + "name": "Ministral 8B (latest)", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-01", + "last_updated": "2024-10-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.1, + "output": 0.1 } }, - "v0-1.5-lg": { - "id": "v0-1.5-lg", - "name": "v0 1.5 LG", + "mistral/codestral": { + "id": "mistral/codestral", + "name": "Codestral (latest)", + "description": "Mistral code model for completions, refactors, and developer IDE workflows", + "family": "codestral", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-04", - "last_updated": "2025-07-04", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-05-29", + "last_updated": "2025-01-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 + "context": 256000, + "output": 4096 }, "cost": { - "input": 15, - "output": 75 + "input": 0.3, + "output": 0.9 } }, - "ernie-4.5-turbo-128k": { - "id": "ernie-4.5-turbo-128k", - "name": "Ernie 4.5 Turbo 128k", + "mistral/pixtral-12b": { + "id": "mistral/pixtral-12b", + "name": "Pixtral 12B", + "description": "Mistral vision-language model for image understanding and multimodal chat", + "family": "pixtral", "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "tool_call": true, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2024-09-01", + "last_updated": "2024-09-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "input": 128000, - "output": 16384 + "output": 128000 }, "cost": { - "input": 0.132, - "output": 0.55 + "input": 0.15, + "output": 0.15 } }, - "KAT-Coder-Pro-V1": { - "id": "KAT-Coder-Pro-V1", - "name": "KAT Coder Pro V1", - "attachment": false, + "mistral/pixtral-large": { + "id": "mistral/pixtral-large", + "name": "Pixtral Large (latest)", + "description": "Mistral's larger vision model for document-heavy image understanding and chat", + "family": "pixtral", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2024-11-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 128000, + "output": 128000 }, "cost": { - "input": 1.5, + "input": 2, "output": 6 } }, - "claude-3-5-sonnet-20240620": { - "id": "claude-3-5-sonnet-20240620", - "name": "Claude 3.5 Sonnet Old", - "attachment": true, + "mistral/ministral-3b": { + "id": "mistral/ministral-3b", + "name": "Ministral 3B (latest)", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2024-06-20", - "last_updated": "2024-06-20", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-01", + "last_updated": "2024-10-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 8192 + "context": 128000, + "output": 128000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.04, + "output": 0.04 } }, - "claude-opus-4-1-thinking:8192": { - "id": "claude-opus-4-1-thinking:8192", - "name": "Claude 4.1 Opus Thinking (8K)", - "attachment": true, + "mistral/magistral-small": { + "id": "mistral/magistral-small", + "name": "Magistral Small", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-small", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-03-17", + "last_updated": "2025-03-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.5, + "output": 1.5 } }, - "gemini-2.0-flash-exp-image-generation": { - "id": "gemini-2.0-flash-exp-image-generation", - "name": "Gemini Text + Image", + "mistral/magistral-medium": { + "id": "mistral/magistral-medium", + "name": "Magistral Medium (latest)", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-medium", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-03-17", + "last_updated": "2025-03-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32767, - "input": 32767, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 2, + "output": 5 } }, - "Llama-3.3-70B-Magnum-v4-SE": { - "id": "Llama-3.3-70B-Magnum-v4-SE", - "name": "Llama 3.3 70B Magnum v4 SE", + "google/gemini-embedding-2": { + "id": "google/gemini-embedding-2", + "name": "Gemini Embedding 2", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "gemini-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-03-10", + "last_updated": "2026-03-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "glm-zero-preview": { - "id": "glm-zero-preview", - "name": "GLM Zero Preview", + "google/text-multilingual-embedding-002": { + "id": "google/text-multilingual-embedding-002", + "name": "Text Multilingual Embedding 002", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "temperature": true, + "release_date": "2024-03-01", + "last_updated": "2024-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 4096 + "context": 8192, + "output": 1536 + } + }, + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65000 }, "cost": { - "input": 1.802, - "output": 1.802 + "input": 0.25, + "output": 1.5, + "cache_read": 0.03 } }, - "study_gpt-chatgpt-4o-latest": { - "id": "study_gpt-chatgpt-4o-latest", - "name": "Study Mode", + "google/gemini-3.1-flash-image": { + "id": "google/gemini-3.1-flash-image", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 4.998, - "output": 14.994 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "glm-4-airx": { - "id": "glm-4-airx", - "name": "GLM-4 AirX", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-06-05", - "last_updated": "2024-06-05", + "google/gemini-3-flash": { + "id": "google/gemini-3-flash", + "name": "Gemini 3 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 4096 + "context": 1000000, + "output": 65000 }, "cost": { - "input": 2.006, - "output": 2.006 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "step-2-mini": { - "id": "step-2-mini", - "name": "Step-2 Mini", + "google/veo-3.1-generate-001": { + "id": "google/veo-3.1-generate-001", + "name": "Veo 3.1", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-05", - "last_updated": "2024-07-05", + "temperature": true, + "release_date": "2025-10-15", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 4096 - }, - "cost": { - "input": 0.2006, - "output": 0.408 + "context": 0, + "output": 0 } }, - "gemini-2.5-flash-preview-04-17:thinking": { - "id": "gemini-2.5-flash-preview-04-17:thinking", - "name": "Gemini 2.5 Flash Preview Thinking", + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-17", - "last_updated": "2025-04-17", + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 3.5 + "input": 1.5, + "output": 9, + "cache_read": 0.15 } }, - "Llama-3.3-70B-Mokume-Gane-R1": { - "id": "Llama-3.3-70B-Mokume-Gane-R1", - "name": "Llama 3.3 70B Mokume Gane R1", - "attachment": false, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.14, + "output": 0.4 } }, - "deepseek-reasoner": { - "id": "deepseek-reasoner", - "name": "DeepSeek Reasoner", + "google/veo-3.0-generate-001": { + "id": "google/veo-3.0-generate-001", + "name": "Veo 3.0", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "temperature": true, + "release_date": "2025-05-20", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 64000, - "input": 64000, - "output": 65536 - }, - "cost": { - "input": 0.4, - "output": 1.7 + "context": 0, + "output": 0 } }, - "glm-z1-airx": { - "id": "glm-z1-airx", - "name": "GLM Z1 AirX", + "google/veo-3.0-fast-generate-001": { + "id": "google/veo-3.0-fast-generate-001", + "name": "Veo 3.0 Fast Generate", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "tool_call": false, + "temperature": true, + "release_date": "2025-07-31", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 16384 - }, - "cost": { - "input": 0.7, - "output": 0.7 + "context": 0, + "output": 0 } }, - "jamba-mini-1.6": { - "id": "jamba-mini-1.6", - "name": "Jamba Mini 1.6", + "google/text-embedding-005": { + "id": "google/text-embedding-005", + "name": "Text Embedding 005", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "temperature": true, + "release_date": "2024-08-01", + "last_updated": "2024-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 - }, - "cost": { - "input": 0.1989, - "output": 0.408 - } - }, - "claude-opus-4-1-thinking": { - "id": "claude-opus-4-1-thinking", - "name": "Claude 4.1 Opus Thinking", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "input": 200000, - "output": 32000 - }, - "cost": { - "input": 14.994, - "output": 75.004 + "context": 8192, + "output": 1536 } }, - "grok-3-beta": { - "id": "grok-3-beta", - "name": "Grok 3 Beta", + "google/gemini-embedding-001": { + "id": "google/gemini-embedding-001", + "name": "Gemini Embedding 001", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "gemini-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 131072 - }, - "cost": { - "input": 3, - "output": 15 + "context": 8192, + "output": 1536 } }, - "Llama-3.3-70B-Legion-V2.1": { - "id": "Llama-3.3-70B-Legion-V2.1", - "name": "Llama 3.3 70B Legion V2.1", - "attachment": false, - "reasoning": false, + "google/gemini-3.1-flash-lite-image": { + "id": "google/gemini-3.1-flash-lite-image", + "name": "Gemini 3.1 Flash Lite Image (Nano Banana 2 Lite)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 65536, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.25, + "output": 1.5, + "cache_read": 0.03 } }, - "sonar": { - "id": "sonar", - "name": "Perplexity Simple", + "google/gemini-2.5-flash-image": { + "id": "google/gemini-2.5-flash-image", + "name": "Nano Banana (Gemini 2.5 Flash Image)", + "description": "Nano Banana image model for fast generation, edits, and character-consistent assets", + "family": "gemini-flash", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 127000, - "input": 127000, - "output": 128000 + "context": 32768, + "output": 65536 }, "cost": { - "input": 1.003, - "output": 1.003 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03 } }, - "z-image-turbo": { - "id": "z-image-turbo", - "name": "Z Image Turbo", - "attachment": true, + "google/imagen-4.0-fast-generate-001": { + "id": "google/imagen-4.0-fast-generate-001", + "name": "Imagen 4 Fast", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, "temperature": true, - "release_date": "2025-11-27", - "last_updated": "2025-11-27", + "release_date": "2025-06-01", + "last_updated": "2025-06", "modalities": { "input": ["text"], "output": ["image"] }, "open_weights": false, "limit": { - "context": 0, + "context": 480, "output": 0 } }, - "GLM-4.5-Air-Derestricted-Iceblink-v2": { - "id": "GLM-4.5-Air-Derestricted-Iceblink-v2", - "name": "GLM 4.5 Air Derestricted Iceblink v2", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 158600, - "input": 158600, + "context": 1048576, "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.1, + "output": 0.4, + "cache_read": 0.01 } }, - "jamba-large": { - "id": "jamba-large", - "name": "Jamba Large", - "attachment": false, - "reasoning": false, + "google/gemini-omni-flash-preview": { + "id": "google/gemini-omni-flash-preview", + "name": "Gemini Omni Flash Preview", + "description": "Omni-modal model for text, vision, audio, and multimodal agent tasks", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "temperature": true, + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 + "context": 1000000, + "output": 57920 }, "cost": { - "input": 1.989, - "output": 7.99 + "input": 1.5, + "output": 9 } }, - "claude-3-7-sonnet-reasoner": { - "id": "claude-3-7-sonnet-reasoner", - "name": "Claude 3.7 Sonnet Reasoner", + "google/gemini-3.1-flash-image-preview": { + "id": "google/gemini-3.1-flash-image-preview", + "name": "Gemini 3.1 Flash Image Preview (Nano Banana 2)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2025-03-29", - "last_updated": "2025-03-29", - "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 3, - "output": 15 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "ernie-4.5-turbo-vl-32k": { - "id": "ernie-4.5-turbo-vl-32k", - "name": "Ernie 4.5 Turbo VL 32k", - "attachment": true, + "google/imagen-4.0-ultra-generate-001": { + "id": "google/imagen-4.0-ultra-generate-001", + "name": "Imagen 4 Ultra", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "temperature": false, + "release_date": "2025-05-24", + "last_updated": "2025-05-24", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 16384 - }, - "cost": { - "input": 0.495, - "output": 1.43 + "context": 480, + "output": 0 } }, - "Mistral-Nemo-12B-Instruct-2407": { - "id": "Mistral-Nemo-12B-Instruct-2407", - "name": "Mistral Nemo 12B Instruct 2407", + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.01, - "output": 0.01 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "doubao-seed-1-6-flash-250615": { - "id": "doubao-seed-1-6-flash-250615", - "name": "Doubao Seed 1.6 Flash", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-15", - "last_updated": "2025-06-15", + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.0374, - "output": 0.374 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "qwq-32b": { - "id": "qwq-32b", - "name": "Qwen: QwQ 32B", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "google/gemini-3-pro-preview": { + "id": "google/gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.25599999, - "output": 0.30499999 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "Llama-3.3-70B-Strawberrylemonade-v1.2": { - "id": "Llama-3.3-70B-Strawberrylemonade-v1.2", - "name": "Llama 3.3 70B StrawberryLemonade v1.2", + "google/veo-3.1-fast-generate-001": { + "id": "google/veo-3.1-fast-generate-001", + "name": "Veo 3.1 Fast Generate", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-10-15", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "gemini-2.5-flash-preview-04-17": { - "id": "gemini-2.5-flash-preview-04-17", - "name": "Gemini 2.5 Flash Preview", - "attachment": true, - "reasoning": true, + "google/gemini-3-pro-image": { + "id": "google/gemini-3-pro-image", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-17", - "last_updated": "2025-04-17", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 65536, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "ernie-x1-turbo-32k": { - "id": "ernie-x1-turbo-32k", - "name": "Ernie X1 Turbo 32k", + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 16384 + "context": 1000000, + "output": 65000 }, "cost": { - "input": 0.165, - "output": 0.66 + "input": 0.25, + "output": 1.5, + "cache_read": 0.03 } }, - "deepseek-math-v2": { - "id": "deepseek-math-v2", - "name": "DeepSeek Math V2", + "google/imagen-4.0-generate-001": { + "id": "google/imagen-4.0-generate-001", + "name": "Imagen 4", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "temperature": true, + "release_date": "2025-05-20", + "last_updated": "2025-05-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 - }, - "cost": { - "input": 0.6, - "output": 2.2 + "context": 480, + "output": 0 } }, - "Llama-3.3-70B-Electranova-v1.0": { - "id": "Llama-3.3-70B-Electranova-v1.0", - "name": "Llama 3.3 70B Electranova v1.0", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 } }, - "Llama-3.3-70B-ArliAI-RPMax-v2": { - "id": "Llama-3.3-70B-ArliAI-RPMax-v2", - "name": "Llama 3.3 70B ArliAI RPMax v2", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "qwen-image": { - "id": "qwen-image", - "name": "Qwen Image", - "attachment": true, + "prodia/flux-fast-schnell": { + "id": "prodia/flux-fast-schnell", + "name": "Flux Schnell", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, "temperature": true, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2024-08-02", + "last_updated": "2026-06-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["image"] }, "open_weights": false, "limit": { - "context": 0, + "context": 512, "output": 0 } }, - "Llama-3.3-70B-Cu-Mai-R1": { - "id": "Llama-3.3-70B-Cu-Mai-R1", - "name": "Llama 3.3 70B Cu Mai R1", + "openai/gpt-oss-safeguard-20b": { + "id": "openai/gpt-oss-safeguard-20b", + "name": "gpt-oss-safeguard-20b", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-10-29", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 131072, + "input": 65536, + "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.075, + "output": 0.3, + "cache_read": 0.037 } }, - "GLM-4.5-Air-Derestricted-Iceblink": { - "id": "GLM-4.5-Air-Derestricted-Iceblink", - "name": "GLM 4.5 Air Derestricted Iceblink", + "openai/gpt-3.5-turbo-instruct": { + "id": "openai/gpt-3.5-turbo-instruct", + "name": "GPT-3.5 Turbo Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "temperature": true, + "knowledge": "2021-09", + "release_date": "2023-09-18", + "last_updated": "2023-03-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 98304 + "context": 8192, + "input": 4096, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.5, + "output": 2 } }, - "Llama-3.3-70B-Bigger-Body": { - "id": "Llama-3.3-70B-Bigger-Body", - "name": "Llama 3.3 70B Bigger Body", - "attachment": false, + "openai/gpt-5.2-chat": { + "id": "openai/gpt-5.2-chat", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-11", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, + "context": 128000, + "input": 111616, "output": 16384 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Llama-3.3+(3.1v3.3)-70B-Hanami-x1": { - "id": "Llama-3.3+(3.1v3.3)-70B-Hanami-x1", - "name": "Llama 3.3+ 70B Hanami x1", + "openai/text-embedding-3-large": { + "id": "openai/text-embedding-3-large", + "name": "text-embedding-3-large", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 8192, + "input": 6656, + "output": 1536 + } + }, + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT 5.2 ", + "description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 21, + "output": 168 } }, - "hunyuan-turbos-20250226": { - "id": "hunyuan-turbos-20250226", - "name": "Hunyuan Turbo S", + "openai/gpt-4o-mini-search-preview": { + "id": "openai/gpt-4o-mini-search-preview", + "name": "GPT 4o Mini Search Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "temperature": true, + "knowledge": "2023-09", + "release_date": "2025-03-12", + "last_updated": "2025-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 24000, - "input": 24000, - "output": 8192 + "context": 128000, + "input": 111616, + "output": 16384 }, "cost": { - "input": 0.187, - "output": 0.374 + "input": 0.15, + "output": 0.6 } }, - "gemini-2.5-flash-preview-09-2025": { - "id": "gemini-2.5-flash-preview-09-2025", - "name": "Gemini 2.5 Flash Preview (09/2025)", + "openai/gpt-5-chat": { + "id": "openai/gpt-5-chat", + "name": "GPT-5 Chat", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image", "pdf"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 128000, + "input": 111616, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "GLM-4.6-Derestricted-v5": { - "id": "GLM-4.6-Derestricted-v5", - "name": "GLM 4.6 Derestricted v5", + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "temperature": true, + "knowledge": "2021-09", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 + "context": 16385, + "input": 12289, + "output": 4096 }, "cost": { - "input": 0.4, + "input": 0.5, "output": 1.5 } }, - "glm-4-plus": { - "id": "glm-4-plus", - "name": "GLM-4 Plus", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-08-01", - "last_updated": "2024-08-01", + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT-5 pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 400000, "input": 128000, - "output": 4096 + "output": 272000 }, "cost": { - "input": 7.497, - "output": 7.497 + "input": 15, + "output": 120 } }, - "Gemma-3-27B-Big-Tiger-v3": { - "id": "Gemma-3-27B-Big-Tiger-v3", - "name": "Gemma 3 27B Big Tiger v3", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "openai/o3-pro": { + "id": "openai/o3-pro", + "name": "o3 Pro", + "description": "High-effort o3 tier for difficult technical reasoning and careful answers", + "family": "o-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-06-10", + "last_updated": "2025-06-10", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 200000, + "input": 100000, + "output": 100000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 20, + "output": 80 } }, - "brave-research": { - "id": "brave-research", - "name": "Brave (Research)", + "openai/gpt-4o-transcribe": { + "id": "openai/gpt-4o-transcribe", + "name": "GPT-4o Transcribe", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2023-03-02", - "last_updated": "2024-01-01", + "temperature": true, + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 0, + "output": 0 }, "cost": { - "input": 5, - "output": 5 + "input": 2.5, + "output": 10 } }, - "hidream": { - "id": "hidream", - "name": "Hidream", + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT 5.4 Nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2024-01-01", + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "qwen3-max-2026-01-23": { - "id": "qwen3-max-2026-01-23", - "name": "Qwen3 Max 2026-01-23", - "attachment": false, + "openai/gpt-5.3-chat": { + "id": "openai/gpt-5.3-chat", + "name": "GPT-5.3 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-26", - "last_updated": "2026-01-26", + "tool_call": true, + "temperature": true, + "release_date": "2026-03-03", + "last_updated": "2026-03-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 128000, + "input": 111616, + "output": 16384 }, "cost": { - "input": 1.2002, - "output": 6.001 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "Claude 4.1 Opus", + "openai/gpt-5.1-thinking": { + "id": "openai/gpt-5.1-thinking", + "name": "GPT 5.1 Thinking", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-12", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image", "pdf"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Claude Haiku 4.5", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1-Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-12", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1, - "output": 5 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "MiniMax-M1": { - "id": "MiniMax-M1", - "name": "MiniMax M1", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-16", - "last_updated": "2025-06-16", + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT 5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-19", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.1394, - "output": 1.3328 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "gemini-2.5-flash-nothinking": { - "id": "gemini-2.5-flash-nothinking", - "name": "Gemini 2.5 Flash (No Thinking)", - "attachment": true, + "openai/gpt-4o-mini-transcribe": { + "id": "openai/gpt-4o-mini-transcribe", + "name": "GPT-4o mini Transcribe", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "o-mini", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "temperature": true, + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 0, + "output": 0 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 1.25, + "output": 5 } }, - "exa-research-pro": { - "id": "exa-research-pro", - "name": "Exa (Research Pro)", + "openai/text-embedding-ada-002": { + "id": "openai/text-embedding-ada-002", + "name": "text-embedding-ada-002", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-04", - "last_updated": "2025-06-04", + "temperature": true, + "release_date": "2022-12-15", + "last_updated": "2022-12-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 2.5, - "output": 2.5 + "context": 8192, + "input": 6656, + "output": 1536 } }, - "grok-3-fast-beta": { - "id": "grok-3-fast-beta", - "name": "Grok 3 Fast Beta", + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "claude-opus-4-5-20251101:thinking": { - "id": "claude-opus-4-5-20251101:thinking", - "name": "Claude 4.5 Opus Thinking", + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT 5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "gemini-2.5-pro-exp-03-25": { - "id": "gemini-2.5-pro-exp-03-25", - "name": "Gemini 2.5 Pro Experimental 0325", - "attachment": true, - "reasoning": true, + "openai/text-embedding-3-small": { + "id": "openai/text-embedding-3-small", + "name": "text-embedding-3-small", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "temperature": true, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 - }, - "cost": { - "input": 2.5, - "output": 10 + "context": 8192, + "input": 6656, + "output": 1536 } }, - "claude-3-7-sonnet-thinking": { - "id": "claude-3-7-sonnet-thinking", - "name": "Claude 3.7 Sonnet Thinking", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "openai/gpt-realtime-1.5": { + "id": "openai/gpt-realtime-1.5", + "name": "GPT-Realtime-1.5", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 16000 + "context": 0, + "output": 0 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 4, + "output": 16, + "cache_read": 0.4 } }, - "claude-opus-4-thinking:8192": { - "id": "claude-opus-4-thinking:8192", - "name": "Claude 4 Opus Thinking (8K)", + "openai/gpt-5.6-luna": { + "id": "openai/gpt-5.6-luna", + "name": "GPT 5.6 Luna", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "claude-sonnet-4-thinking:1024": { - "id": "claude-sonnet-4-thinking:1024", - "name": "Claude 4 Sonnet Thinking (1K)", + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-12", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "Llama-3.3-70B-Magnum-v4-SE-Cirrus-x1-SLERP": { - "id": "Llama-3.3-70B-Magnum-v4-SE-Cirrus-x1-SLERP", - "name": "Llama 3.3 70B Magnum v4 SE Cirrus x1 SLERP", + "openai/gpt-realtime-2.1": { + "id": "openai/gpt-realtime-2.1", + "name": "gpt-realtime-2.1", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high", "xhigh"] + } + ], "tool_call": false, "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "temperature": true, + "knowledge": "2024-09-30", + "release_date": "2026-07-09", + "last_updated": "2026-07-06", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "input": 96000, + "output": 32000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 4, + "output": 24, + "cache_read": 0.4 } }, - "step-r1-v-mini": { - "id": "step-r1-v-mini", - "name": "Step R1 V Mini", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-08", - "last_updated": "2025-04-08", + "openai/gpt-5.6-terra": { + "id": "openai/gpt-5.6-terra", + "name": "GPT 5.6 Terra", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { "input": 2.5, - "output": 11 - } - }, - "ernie-x1-32k-preview": { - "id": "ernie-x1-32k-preview", - "name": "Ernie X1 32k", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 32000, - "input": 32000, - "output": 16384 - }, - "cost": { - "input": 0.33, - "output": 1.32 + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "Llama-3.3-70B-StrawberryLemonade-v1.0": { - "id": "Llama-3.3-70B-StrawberryLemonade-v1.0", - "name": "Llama 3.3 70B StrawberryLemonade v1.0", + "openai/gpt-image-1.5": { + "id": "openai/gpt-image-1.5", + "name": "GPT Image 1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 0, + "output": 0 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 5, + "output": 32, + "cache_read": 1.25 } }, - "KAT-Coder-Exp-72B-1010": { - "id": "KAT-Coder-Exp-72B-1010", - "name": "KAT Coder Exp 72B 1010", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { "input": 0.1, - "output": 0.2 + "output": 0.5 } }, - "gemini-2.5-pro-preview-03-25": { - "id": "gemini-2.5-pro-preview-03-25", - "name": "Gemini 2.5 Pro Preview 0325", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT 5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { "input": 2.5, - "output": 10 + "output": 15, + "cache_read": 0.25 } }, - "claude-opus-4-thinking:1024": { - "id": "claude-opus-4-thinking:1024", - "name": "Claude 4 Opus Thinking (1K)", + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT 5.4 Mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "Claude 4 Sonnet", - "attachment": true, + "openai/gpt-realtime-mini": { + "id": "openai/gpt-realtime-mini", + "name": "GPT-Realtime mini", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "tool_call": false, + "temperature": true, + "release_date": "2025-10-10", + "last_updated": "2025-10-10", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 0, + "output": 0 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06 } }, - "Llama-3.3-70B-Progenitor-V3.3": { - "id": "Llama-3.3-70B-Progenitor-V3.3", - "name": "Llama 3.3 70B Progenitor V3.3", + "openai/tts-1": { + "id": "openai/tts-1", + "name": "TTS-1", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "temperature": true, + "release_date": "2023-11-06", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 0, + "output": 0 } }, - "Qwen2.5-32B-EVA-v0.2": { - "id": "Qwen2.5-32B-EVA-v0.2", - "name": "Qwen 2.5 32b EVA", + "openai/whisper-1": { + "id": "openai/whisper-1", + "name": "Whisper", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-09-01", - "last_updated": "2024-09-01", + "temperature": true, + "release_date": "2022-09-21", + "last_updated": "2022-09-21", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 24576, - "input": 24576, - "output": 8192 - }, - "cost": { - "input": 0.493, - "output": 0.493 + "context": 0, + "output": 0 } }, - "brave-pro": { - "id": "brave-pro", - "name": "Brave (Pro)", + "openai/tts-1-hd": { + "id": "openai/tts-1-hd", + "name": "TTS-1 HD", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2023-03-02", - "last_updated": "2024-01-01", + "temperature": true, + "release_date": "2023-11-06", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 - }, - "cost": { - "input": 5, - "output": 5 + "context": 0, + "output": 0 } }, - "step-2-16k-exp": { - "id": "step-2-16k-exp", - "name": "Step-2 16k Exp", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-05", - "last_updated": "2024-07-05", + "openai/o3-deep-research": { + "id": "openai/o3-deep-research", + "name": "o3-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-06-26", + "last_updated": "2024-06-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16000, - "input": 16000, - "output": 8192 + "context": 200000, + "input": 100000, + "output": 100000 }, "cost": { - "input": 7.004, - "output": 19.992 + "input": 10, + "output": 40, + "cache_read": 2.5 } }, - "Llama-3.3-70B-Fallen-R1-v1": { - "id": "Llama-3.3-70B-Fallen-R1-v1", - "name": "Llama 3.3 70B Fallen R1 v1", + "openai/gpt-image-1": { + "id": "openai/gpt-image-1", + "name": "GPT Image 1", + "description": "OpenAI image model for production generation, edits, and brand-safe visual workflows", + "family": "gpt-image", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-03-25", + "last_updated": "2025-04-24", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 0, + "output": 0 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 5, + "output": 40, + "cache_read": 1.25 } }, - "claude-sonnet-4-thinking": { - "id": "claude-sonnet-4-thinking", - "name": "Claude 4 Sonnet Thinking", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "openai/gpt-realtime-2": { + "id": "openai/gpt-realtime-2", + "name": "gpt-realtime-2", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 + "context": 0, + "output": 0 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 4, + "output": 24, + "cache_read": 0.4 } }, - "doubao-1.5-pro-256k": { - "id": "doubao-1.5-pro-256k", - "name": "Doubao 1.5 Pro 256k", + "openai/gpt-image-1-mini": { + "id": "openai/gpt-image-1-mini", + "name": "GPT Image 1 Mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "temperature": true, + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 0, + "output": 0 }, "cost": { - "input": 0.799, - "output": 1.445 + "input": 2, + "output": 8, + "cache_read": 0.2 } }, - "claude-3-7-sonnet-20250219": { - "id": "claude-3-7-sonnet-20250219", - "name": "Claude 3.7 Sonnet", + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT 5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "structured_output": false, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 16000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 30, + "output": 180 } }, - "learnlm-1.5-pro-experimental": { - "id": "learnlm-1.5-pro-experimental", - "name": "Gemini LearnLM Experimental", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-05-14", - "last_updated": "2024-05-14", + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT 5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-12-01", + "release_date": "2026-04-24", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32767, - "input": 32767, - "output": 8192 + "context": 1000000, + "input": 872000, + "output": 128000 }, "cost": { - "input": 3.502, - "output": 10.506 + "input": 30, + "output": 180 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3 Coder 30B A3B Instruct", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, + "temperature": true, + "knowledge": "2024-10", "release_date": "2025-08-05", "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 131072, + "input": 122880, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.05, + "output": 0.2 } }, - "chroma": { - "id": "chroma", - "name": "Chroma", + "openai/gpt-5.6-sol": { + "id": "openai/gpt-5.6-sol", + "name": "GPT 5.6 Sol", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-12", - "last_updated": "2025-08-12", - "modalities": { - "input": ["text"], - "output": ["image"] - }, - "open_weights": false, - "limit": { - "context": 0, - "output": 0 - } - }, - "Llama-3.3-70B-Predatorial-Extasy": { - "id": "Llama-3.3-70B-Predatorial-Extasy", - "name": "Llama 3.3 70B Predatorial Extasy", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "Llama-3.3-70B-Aurora-Borealis": { - "id": "Llama-3.3-70B-Aurora-Borealis", - "name": "Llama 3.3 70B Aurora Borealis", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2-Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-18", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Llama-3.3-70B-ArliAI-RPMax-v3": { - "id": "Llama-3.3-70B-ArliAI-RPMax-v3", - "name": "Llama 3.3 70B ArliAI RPMax v3", - "attachment": false, + "openai/gpt-5.1-instant": { + "id": "openai/gpt-5.1-instant", + "name": "GPT-5.1 Instant", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-12", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, + "context": 128000, + "input": 111616, "output": 16384 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "venice-uncensored": { - "id": "venice-uncensored", - "name": "Venice Uncensored", + "openai/gpt-image-2": { + "id": "openai/gpt-image-2", + "name": "GPT Image 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 0, + "output": 0 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 5, + "output": 30, + "cache_read": 1.25 } }, - "step-3": { - "id": "step-3", - "name": "Step-3", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT 5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-12-01", + "release_date": "2026-04-24", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 + "context": 1000000, + "input": 872000, + "output": 128000 }, "cost": { - "input": 0.2499, - "output": 0.6494 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "Llama-3.3-70B-The-Omega-Directive-Unslop-v2.0": { - "id": "Llama-3.3-70B-The-Omega-Directive-Unslop-v2.0", - "name": "Llama 3.3 70B Omega Directive Unslop v2.0", + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "auto-model": { - "id": "auto-model", - "name": "Auto model", - "attachment": false, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 1000000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "claude-opus-4-1-thinking:32768": { - "id": "claude-opus-4-1-thinking:32768", - "name": "Claude 4.1 Opus Thinking (32K)", + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "Llama-3.3-70B-Shakudo": { - "id": "Llama-3.3-70B-Shakudo", - "name": "Llama 3.3 70B Shakudo", - "attachment": false, + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 10, + "output": 30 } }, - "Baichuan4-Air": { - "id": "Baichuan4-Air", - "name": "Baichuan 4 Air", - "attachment": false, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, + "context": 1047576, "output": 32768 }, "cost": { - "input": 0.157, - "output": 0.157 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "kimi-thinking-preview": { - "id": "kimi-thinking-preview", - "name": "Kimi Thinking Preview", + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-07", - "last_updated": "2025-05-07", - "modalities": { - "input": ["text", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 31.46, - "output": 31.46 - } - }, - "qwen-turbo": { - "id": "qwen-turbo", - "name": "Qwen Turbo", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.04998, - "output": 0.2006 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "Llama-3.3-70B-Mhnnn-x1": { - "id": "Llama-3.3-70B-Mhnnn-x1", - "name": "Llama 3.3 70B Mhnnn x1", - "attachment": false, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "claude-opus-4-thinking:32768": { - "id": "claude-opus-4-thinking:32768", - "name": "Claude 4 Opus Thinking (32K)", + "openai/o1": { + "id": "openai/o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -9175,887 +10607,1103 @@ "open_weights": false, "limit": { "context": 200000, - "input": 200000, - "output": 32000 + "output": 100000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "Llama-3.3-70B-Argunaut-1-SFT": { - "id": "Llama-3.3-70B-Argunaut-1-SFT", - "name": "Llama 3.3 70B Argunaut 1 SFT", - "attachment": false, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "claude-opus-4-1-thinking:1024": { - "id": "claude-opus-4-1-thinking:1024", - "name": "Claude 4.1 Opus Thinking (1K)", - "attachment": true, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "input": 200000, - "output": 32000 + "output": 100000 }, "cost": { - "input": 14.994, - "output": 75.004 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "phi-4-multimodal-instruct": { - "id": "phi-4-multimodal-instruct", - "name": "Phi 4 Multimodal", - "attachment": false, + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "input": 128000, "output": 16384 }, "cost": { - "input": 0.07, - "output": 0.11 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "doubao-seed-2-0-code-preview-260215": { - "id": "doubao-seed-2-0-code-preview-260215", - "name": "Doubao Seed 2.0 Code Preview", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, + "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.782, - "output": 3.893 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "deepseek-reasoner-cheaper": { - "id": "deepseek-reasoner-cheaper", - "name": "Deepseek R1 Cheaper", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.4, - "output": 1.7 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "exa-answer": { - "id": "exa-answer", - "name": "Exa (Answer)", + "zai/glm-4.7": { + "id": "zai/glm-4.7", + "name": "GLM 4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-04", - "last_updated": "2025-06-04", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "input": 4096, - "output": 4096 + "context": 200000, + "output": 120000 }, "cost": { - "input": 2.5, - "output": 2.5 + "input": 0.6, + "output": 2.2, + "cache_read": 0.12 } }, - "v0-1.0-md": { - "id": "v0-1.0-md", - "name": "v0 1.0 MD", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-04", - "last_updated": "2025-07-04", + "zai/glm-4.5v": { + "id": "zai/glm-4.5v", + "name": "GLM 4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 + "context": 66000, + "output": 16000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.6, + "output": 1.8, + "cache_read": 0.11 } }, - "glm-4.1v-thinking-flash": { - "id": "glm-4.1v-thinking-flash", - "name": "GLM 4.1V Thinking Flash", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "zai/glm-4.5": { + "id": "zai/glm-4.5", + "name": "GLM 4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 64000, - "input": 64000, - "output": 8192 + "context": 128000, + "output": 96000 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "azure-o1": { - "id": "azure-o1", - "name": "Azure o1", + "zai/glm-4.7-flashx": { + "id": "zai/glm-4.7-flashx", + "name": "GLM 4.7 FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-17", - "last_updated": "2024-12-17", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 200000, - "input": 200000, - "output": 100000 + "output": 128000 }, "cost": { - "input": 14.994, - "output": 59.993 + "input": 0.06, + "output": 0.4, + "cache_read": 0.01 } }, - "GLM-4.5-Air-Derestricted": { - "id": "GLM-4.5-Air-Derestricted", - "name": "GLM 4.5 Air Derestricted", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "zai/glm-5.1": { + "id": "zai/glm-5.1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 202600, - "input": 202600, - "output": 98304 + "context": 202000, + "output": 202000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.3, + "output": 4.3, + "cache_read": 0.26 } }, - "azure-o3-mini": { - "id": "azure-o3-mini", - "name": "Azure o3-mini", + "zai/glm-4.6": { + "id": "zai/glm-4.6", + "name": "GLM 4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 200000, - "input": 200000, - "output": 65536 + "output": 96000 }, "cost": { - "input": 1.088, - "output": 4.3996 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "qwen3.6-max-preview": { - "id": "qwen3.6-max-preview", - "name": "Qwen3.6 Max Preview", - "family": "qwen3.6", + "zai/glm-5.2": { + "id": "zai/glm-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-04-20", - "last_updated": "2026-04-21", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-16", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 245800, - "output": 65536 + "context": 1040000, + "output": 128000 }, "cost": { - "input": 1.3, - "output": 7.8 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "Llama-3.3-70B-Sapphira-0.2": { - "id": "Llama-3.3-70B-Sapphira-0.2", - "name": "Llama 3.3 70B Sapphira 0.2", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "zai/glm-4.6v": { + "id": "zai/glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-30", + "last_updated": "2025-12-08", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 128000, + "output": 24000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.3, + "output": 0.9, + "cache_read": 0.05 } }, - "Llama-3.3-70B-Anthrobomination": { - "id": "Llama-3.3-70B-Anthrobomination", - "name": "Llama 3.3 70B Anthrobomination", + "zai/glm-5.2-fast": { + "id": "zai/glm-5.2-fast", + "name": "GLM 5.2 Fast", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-23", + "last_updated": "2026-06-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 2.1, + "output": 6.6, + "cache_read": 0.21 } }, - "QwQ-32B-ArliAI-RpR-v1": { - "id": "QwQ-32B-ArliAI-RpR-v1", - "name": "QwQ 32b Arli V1", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "zai/glm-5v-turbo": { + "id": "zai/glm-5v-turbo", + "name": "GLM 5V Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "Claude 4 Opus", + "zai/glm-4.6v-flash": { + "id": "zai/glm-4.6v-flash", + "name": "GLM-4.6V-Flash", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-05-14", - "last_updated": "2025-05-14", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 32000 - }, - "cost": { - "input": 14.994, - "output": 75.004 + "context": 128000, + "output": 24000 } }, - "yi-lightning": { - "id": "yi-lightning", - "name": "Yi Lightning", + "zai/glm-4.5-air": { + "id": "zai/glm-4.5-air", + "name": "GLM 4.5 Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-10-16", - "last_updated": "2024-10-16", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 12000, - "input": 12000, - "output": 4096 + "context": 128000, + "output": 96000 }, "cost": { - "input": 0.2006, - "output": 0.2006 + "input": 0.2, + "output": 1.1, + "cache_read": 0.03 } }, - "Llama-3.3-70B-Electra-R1": { - "id": "Llama-3.3-70B-Electra-R1", - "name": "Llama 3.3 70B Electra R1", + "zai/glm-4.7-flash": { + "id": "zai/glm-4.7-flash", + "name": "GLM 4.7 Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 200000, + "output": 131000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.07, + "output": 0.4 } }, - "Llama-3.3-70B-Forgotten-Abomination-v5.0": { - "id": "Llama-3.3-70B-Forgotten-Abomination-v5.0", - "name": "Llama 3.3 70B Forgotten Abomination v5.0", + "zai/glm-5": { + "id": "zai/glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 202800, + "output": 131100 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.95, + "output": 3.15, + "cache_read": 0.2 } }, - "Llama-3.3-70B-Cirrus-x1": { - "id": "Llama-3.3-70B-Cirrus-x1", - "name": "Llama 3.3 70B Cirrus x1", + "zai/glm-5-turbo": { + "id": "zai/glm-5-turbo", + "name": "GLM 5 Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 202800, + "output": 131100 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "grok-3-mini-beta": { - "id": "grok-3-mini-beta", - "name": "Grok 3 Mini Beta", + "bytedance/seedream-5.0-lite": { + "id": "bytedance/seedream-5.0-lite", + "name": "Seedream 5.0 Lite", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "seed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-01-28", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 131072 + "context": 0, + "output": 0 + } + }, + "bytedance/seedance-2.0-fast": { + "id": "bytedance/seedance-2.0-fast", + "name": "Seedance 2.0 Fast", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-04-14", + "last_updated": "2026-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["video"] }, - "cost": { - "input": 0.3, - "output": 0.5 + "open_weights": false, + "limit": { + "context": 0, + "output": 0 } }, - "auto-model-standard": { - "id": "auto-model-standard", - "name": "Auto model (Standard)", + "bytedance/seedance-v1.0-pro": { + "id": "bytedance/seedance-v1.0-pro", + "name": "Seedance v1.0 Pro", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "temperature": true, + "release_date": "2025-06-11", + "last_updated": "2025-06-11", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 1000000 - }, - "cost": { - "input": 9.996, - "output": 19.992 + "context": 0, + "output": 0 } }, - "claude-sonnet-4-5-20250929-thinking": { - "id": "claude-sonnet-4-5-20250929-thinking", - "name": "Claude Sonnet 4.5 Thinking", - "attachment": true, + "bytedance/seed-1.6": { + "id": "bytedance/seed-1.6", + "name": "Seed 1.6", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-01", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 64000 + "context": 256000, + "output": 32000 }, "cost": { - "input": 2.992, - "output": 14.994 + "input": 0.25, + "output": 2, + "cache_read": 0.05 } }, - "v0-1.5-md": { - "id": "v0-1.5-md", - "name": "v0 1.5 MD", - "attachment": false, + "bytedance/seedance-2.0": { + "id": "bytedance/seedance-2.0", + "name": "Seedance 2.0", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-04", - "last_updated": "2025-07-04", + "temperature": true, + "release_date": "2026-04-14", + "last_updated": "2026-04-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 + "context": 0, + "output": 0 } }, - "kimi-k2-instruct-fast": { - "id": "kimi-k2-instruct-fast", - "name": "Kimi K2 0711 Fast", - "attachment": true, + "bytedance/seedance-v1.5-pro": { + "id": "bytedance/seedance-v1.5-pro", + "name": "Seedance v1.5 Pro", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-15", - "last_updated": "2025-07-15", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 - }, - "cost": { - "input": 0.1, - "output": 2 + "context": 0, + "output": 0 } }, - "glm-4-long": { - "id": "glm-4-long", - "name": "GLM-4 Long", + "bytedance/seedance-v1.0-pro-fast": { + "id": "bytedance/seedance-v1.0-pro-fast", + "name": "Seedance v1.0 Pro Fast", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-08-01", - "last_updated": "2024-08-01", + "temperature": true, + "release_date": "2025-10-24", + "last_updated": "2025-10-31", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 4096 - }, - "cost": { - "input": 0.2006, - "output": 0.2006 + "context": 0, + "output": 0 } }, - "jamba-large-1.7": { - "id": "jamba-large-1.7", - "name": "Jamba Large 1.7", + "bytedance/seedream-4.0": { + "id": "bytedance/seedream-4.0", + "name": "Seedream 4.0", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "seed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "temperature": true, + "release_date": "2025-09-09", + "last_updated": "2025-08-28", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 4096 - }, - "cost": { - "input": 1.989, - "output": 7.99 + "context": 0, + "output": 0 } }, - "qvq-max": { - "id": "qvq-max", - "name": "Qwen: QvQ Max", - "attachment": true, + "bytedance/seedream-4.5": { + "id": "bytedance/seedream-4.5", + "name": "Seedream 4.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "seed", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-28", - "last_updated": "2025-03-28", + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-11-28", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 - }, - "cost": { - "input": 1.4, - "output": 5.3 + "context": 0, + "output": 0 } }, - "gemini-2.0-flash-thinking-exp-1219": { - "id": "gemini-2.0-flash-thinking-exp-1219", - "name": "Gemini 2.0 Flash Thinking 1219", + "bytedance/seedream-5.0-pro": { + "id": "bytedance/seedream-5.0-pro", + "name": "Seedream 5.0 Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "seed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-19", - "last_updated": "2024-12-19", + "temperature": true, + "release_date": "2026-07-11", + "last_updated": "2026-07-11", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32767, - "input": 32767, - "output": 8192 - }, - "cost": { - "input": 0.1003, - "output": 0.408 + "context": 0, + "output": 0 } }, - "gemini-2.0-flash-lite": { - "id": "gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "bytedance/seed-1.8": { + "id": "bytedance/seed-1.8", + "name": "Seed 1.8", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-01", + "last_updated": "2025-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 8192 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.0748, - "output": 0.306 + "input": 0.25, + "output": 2, + "cache_read": 0.05 } }, - "azure-gpt-4-turbo": { - "id": "azure-gpt-4-turbo", - "name": "Azure gpt-4-turbo", + "morph/morph-v3-large": { + "id": "morph/morph-v3-large", + "name": "Morph v3 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "morph", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2023-11-06", - "last_updated": "2024-01-01", + "temperature": false, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 32000, + "output": 32000 }, "cost": { - "input": 9.996, - "output": 30.005 + "input": 0.9, + "output": 1.9 } }, - "Baichuan-M2": { - "id": "Baichuan-M2", - "name": "Baichuan M2 32B Medical", + "morph/morph-v3-fast": { + "id": "morph/morph-v3-fast", + "name": "Morph v3 Fast", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "morph", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "temperature": false, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 16000, + "output": 16000 }, "cost": { - "input": 15.73, - "output": 15.73 + "input": 0.8, + "output": 1.2 } }, - "qwen-long": { - "id": "qwen-long", - "name": "Qwen Long 10M", + "sakana/fugu-ultra": { + "id": "sakana/fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "aura", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-21", + "last_updated": "2026-06-21", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 10000000, - "input": 10000000, - "output": 8192 + "context": 1000000, + "output": 1000000 }, "cost": { - "input": 0.1003, - "output": 0.408 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "sonar-reasoning-pro": { - "id": "sonar-reasoning-pro", - "name": "Perplexity Reasoning Pro", + "nvidia/nemotron-3-ultra-550b-a55b": { + "id": "nvidia/nemotron-3-ultra-550b-a55b", + "name": "Nemotron 3 Ultra", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 127000, - "input": 127000, - "output": 128000 + "context": 1000000, + "output": 65000 }, "cost": { - "input": 2.006, - "output": 7.9985 + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 } }, - "gemini-2.5-flash-preview-05-20:thinking": { - "id": "gemini-2.5-flash-preview-05-20:thinking", - "name": "Gemini 2.5 Flash 0520 Thinking", - "attachment": true, + "nvidia/nemotron-3-nano-30b-a3b": { + "id": "nvidia/nemotron-3-nano-30b-a3b", + "name": "Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "input": 1048000, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.15, - "output": 3.5 + "input": 0.05, + "output": 0.24 } }, - "GLM-4.5-Air-Derestricted-Steam-ReExtract": { - "id": "GLM-4.5-Air-Derestricted-Steam-ReExtract", - "name": "GLM 4.5 Air Derestricted Steam ReExtract", + "nvidia/nemotron-nano-9b-v2": { + "id": "nvidia/nemotron-nano-9b-v2", + "name": "Nvidia Nemotron Nano 9B V2", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-08-18", + "last_updated": "2025-08-18", "modalities": { "input": ["text"], "output": ["text"] @@ -10063,670 +11711,870 @@ "open_weights": false, "limit": { "context": 131072, - "input": 131072, - "output": 65536 + "output": 131072 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.06, + "output": 0.23 } }, - "Llama-3.3-70B-Dark-Ages-v0.1": { - "id": "Llama-3.3-70B-Dark-Ages-v0.1", - "name": "Llama 3.3 70B Dark Ages v0.1", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "NVIDIA Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 16384 + "context": 256000, + "output": 32000 }, "cost": { - "input": 0.306, - "output": 0.306 + "input": 0.15, + "output": 0.65 } }, - "Baichuan4-Turbo": { - "id": "Baichuan4-Turbo", - "name": "Baichuan 4 Turbo", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "nvidia/nemotron-nano-12b-v2-vl": { + "id": "nvidia/nemotron-nano-12b-v2-vl", + "name": "Nvidia Nemotron Nano 12B V2 VL", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "family": "nemotron", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2.42, - "output": 2.42 + "input": 0.2, + "output": 0.6 } }, - "doubao-1.5-vision-pro-32k": { - "id": "doubao-1.5-vision-pro-32k", - "name": "Doubao 1.5 Vision Pro 32k", + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "MiMo M2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo-v2.5", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-22", - "last_updated": "2025-01-22", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 8192 + "context": 1050000, + "output": 131100 }, "cost": { - "input": 0.459, - "output": 1.377 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "alibaba/qwen3.6-flash": { - "id": "alibaba/qwen3.6-flash", - "name": "Qwen3.6 Flash", - "family": "qwen3.6", + "xiaomi/mimo-v2-flash": { + "id": "xiaomi/mimo-v2-flash", + "name": "MiMo V2 Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-16", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 991800, - "output": 65536 + "context": 262144, + "output": 32000 }, "cost": { - "input": 0.19, - "output": 1.16 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01 } }, - "inflection/inflection-3-pi": { - "id": "inflection/inflection-3-pi", - "name": "Inflection 3 Pi", - "family": "gpt", + "xiaomi/mimo-v2-pro": { + "id": "xiaomi/mimo-v2-pro", + "name": "MiMo V2 Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-10-11", - "last_updated": "2024-10-11", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2.499, - "output": 9.996 + "input": 1, + "output": 3, + "cache_read": 0.2 } }, - "inflection/inflection-3-productivity": { - "id": "inflection/inflection-3-productivity", - "name": "Inflection 3 Productivity", - "family": "gpt", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-10-11", - "last_updated": "2024-10-11", + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "MiMo V2.5 Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo-v2.5-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "input": 8000, - "output": 4096 + "context": 1050000, + "output": 131000 }, "cost": { - "input": 2.499, - "output": 9.996 + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } }, - "essentialai/rnj-1-instruct": { - "id": "essentialai/rnj-1-instruct", - "name": "RNJ-1 Instruct 8B", - "family": "rnj", + "quiverai/arrow-1.1": { + "id": "quiverai/arrow-1.1", + "name": "Arrow 1.1", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-13", - "last_updated": "2025-12-13", + "temperature": true, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 - }, - "cost": { - "input": 0.15, - "output": 0.15 + "context": 131072, + "output": 131072 } }, - "LLM360/K2-Think": { - "id": "LLM360/K2-Think", - "name": "K2-Think", - "family": "kimi-thinking", + "inception/mercury-coder-small": { + "id": "inception/mercury-coder-small", + "name": "Mercury Coder Small Beta", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "mercury", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "temperature": true, + "release_date": "2025-02-26", + "last_updated": "2025-02-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 32000, + "output": 16384 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.25, + "output": 1 } }, - "TEE/kimi-k2.5": { - "id": "TEE/kimi-k2.5", - "name": "Kimi K2.5 TEE", - "family": "kimi", + "inception/mercury-2": { + "id": "inception/mercury-2", + "name": "Mercury 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "mercury", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", - "modalities": { - "input": ["text"], - "output": ["text"] - }, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-24", + "last_updated": "2026-03-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, "open_weights": false, "limit": { "context": 128000, - "input": 128000, - "output": 65535 + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.9 + "input": 0.25, + "output": 0.75, + "cache_read": 0.024999999999999998 } }, - "TEE/glm-4.7": { - "id": "TEE/glm-4.7", - "name": "GLM 4.7 TEE", - "family": "glm", - "attachment": false, + "anthropic/claude-3.5-haiku": { + "id": "anthropic/claude-3.5-haiku", + "name": "Claude 3.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "tool_call": true, + "temperature": true, + "knowledge": "2024-07-31", + "release_date": "2024-11-04", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "input": 131000, - "output": 65535 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0.85, - "output": 3.3 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "TEE/qwen3.5-397b-a17b": { - "id": "TEE/qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B TEE", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-28", - "last_updated": "2026-02-28", + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 258048, - "input": 258048, - "output": 65536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "TEE/glm-5": { - "id": "TEE/glm-5", - "name": "GLM 5 TEE", - "family": "glm", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 203000, - "input": 203000, - "output": 65535 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.2, - "output": 3.5 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "TEE/qwen2.5-vl-72b-instruct": { - "id": "TEE/qwen2.5-vl-72b-instruct", - "name": "Qwen2.5 VL 72B TEE", - "family": "qwen", + "anthropic/claude-sonnet-5": { + "id": "anthropic/claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-01", - "last_updated": "2025-02-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-01-31", + "release_date": "2026-06-29", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "TEE/minimax-m2.1": { - "id": "TEE/minimax-m2.1", - "name": "MiniMax M2.1 TEE", - "family": "minimax", - "attachment": false, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "temperature": true, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 131072 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "TEE/qwen3-30b-a3b-instruct-2507": { - "id": "TEE/qwen3-30b-a3b-instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507 TEE", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "input": 262000, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.44999999999999996 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "TEE/deepseek-v3.1": { - "id": "TEE/deepseek-v3.1", - "name": "DeepSeek V3.1 TEE", - "family": "deepseek", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "anthropic/claude-fable-5": { + "id": "anthropic/claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-01-31", + "release_date": "2026-07-01", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "input": 164000, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1, - "output": 2.5 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "TEE/llama3-3-70b": { - "id": "TEE/llama3-3-70b", - "name": "Llama 3.3 70B", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-03", - "last_updated": "2025-07-03", + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 2, - "output": 2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "TEE/glm-4.6": { - "id": "TEE/glm-4.6", - "name": "GLM 4.6 TEE", - "family": "glm", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 203000, - "input": 203000, - "output": 65535 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.75, - "output": 2 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "TEE/kimi-k2.5-thinking": { - "id": "TEE/kimi-k2.5-thinking", - "name": "Kimi K2.5 Thinking TEE", - "family": "kimi-thinking", - "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "anthropic/claude-3-haiku": { + "id": "anthropic/claude-3-haiku", + "name": "Claude Haiku 3", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-haiku", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-08-31", + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65535 + "context": 200000, + "output": 4096 }, "cost": { - "input": 0.3, - "output": 1.9 + "input": 0.25, + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 } }, - "TEE/gemma-3-27b-it": { - "id": "TEE/gemma-3-27b-it", - "name": "Gemma 3 27B TEE", - "family": "gemma", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "TEE/deepseek-v3.2": { - "id": "TEE/deepseek-v3.2", - "name": "DeepSeek V3.2 TEE", - "family": "deepseek", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "input": 164000, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 1 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "TEE/gpt-oss-20b": { - "id": "TEE/gpt-oss-20b", - "name": "GPT-OSS 20B TEE", - "family": "gpt-oss", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-03-31", "release_date": "2025-08-05", "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 + "context": 200000, + "output": 32000 }, + "status": "deprecated", "cost": { - "input": 0.2, - "output": 0.8 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "TEE/qwen3-coder": { - "id": "TEE/qwen3-coder", - "name": "Qwen3 Coder 480B TEE", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 1.5, - "output": 2 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "TEE/glm-4.7-flash": { - "id": "TEE/glm-4.7-flash", - "name": "GLM 4.7 Flash TEE", - "family": "glm-flash", + "cohere/rerank-v3.5": { + "id": "cohere/rerank-v3.5", + "name": "Cohere Rerank 3.5", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2024-12-02", + "last_updated": "2024-12-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 203000, - "input": 203000, - "output": 65535 - }, - "cost": { - "input": 0.15, - "output": 0.5 + "context": 4096, + "output": 4096 } }, - "TEE/gpt-oss-120b": { - "id": "TEE/gpt-oss-120b", - "name": "GPT-OSS 120B TEE", - "family": "gpt-oss", + "cohere/command-a": { + "id": "cohere/command-a", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 + "context": 256000, + "output": 8000 }, "cost": { - "input": 2, - "output": 2 + "input": 2.5, + "output": 10 } }, - "TEE/deepseek-r1-0528": { - "id": "TEE/deepseek-r1-0528", - "name": "DeepSeek R1 0528 TEE", - "family": "deepseek", + "cohere/rerank-v4-fast": { + "id": "cohere/rerank-v4-fast", + "name": "Cohere Rerank 4 Fast", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "temperature": true, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 - }, - "cost": { - "input": 2, - "output": 2 + "context": 32000, + "output": 32000 } }, - "TEE/kimi-k2-thinking": { - "id": "TEE/kimi-k2-thinking", - "name": "Kimi K2 Thinking TEE", - "family": "kimi-thinking", + "cohere/embed-v4.0": { + "id": "cohere/embed-v4.0", + "name": "Embed v4.0", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "temperature": true, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] @@ -10734,799 +12582,751 @@ "open_weights": false, "limit": { "context": 128000, - "input": 128000, - "output": 65535 - }, - "cost": { - "input": 2, - "output": 2 + "output": 1536 } }, - "CrucibleLab/L3.3-70B-Loki-V2.0": { - "id": "CrucibleLab/L3.3-70B-Loki-V2.0", - "name": "L3.3 70B Loki v2.0", - "family": "llama", + "cohere/rerank-v4-pro": { + "id": "cohere/rerank-v4-pro", + "name": "Cohere Rerank 4 Pro", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "temperature": true, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 32000, + "output": 32000 } }, - "deepseek/deepseek-v3.2:thinking": { - "id": "deepseek/deepseek-v3.2:thinking", - "name": "DeepSeek V3.2 Thinking", - "family": "deepseek", + "stepfun/step-3.7-flash": { + "id": "stepfun/step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", + "family": "step", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-29", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 163000, - "input": 163000, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "deepseek/deepseek-prover-v2-671b": { - "id": "deepseek/deepseek-prover-v2-671b", - "name": "DeepSeek Prover v2 671B", - "family": "deepseek", + "stepfun/step-3.5-flash": { + "id": "stepfun/step-3.5-flash", + "name": "StepFun 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", + "family": "step", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-30", - "last_updated": "2025-04-30", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-29", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 160000, - "input": 160000, - "output": 16384 + "context": 262114, + "output": 262114 }, "cost": { - "input": 1, - "output": 2.5 + "input": 0.09, + "output": 0.3, + "cache_read": 0.02 } }, - "deepseek/deepseek-v3.2-speciale": { - "id": "deepseek/deepseek-v3.2-speciale", - "name": "DeepSeek V3.2 Speciale", - "family": "deepseek", - "attachment": true, + "interfaze/interfaze-beta": { + "id": "interfaze/interfaze-beta", + "name": "Interfaze Beta", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "temperature": true, + "release_date": "2025-10-07", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 163000, - "input": 163000, - "output": 65536 + "context": 1000000, + "output": 32000 }, "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "input": 1.5, + "output": 3.5 } }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": true, + "bfl/flux-kontext-max": { + "id": "bfl/flux-kontext-max", + "name": "FLUX.1 Kontext Max", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "tool_call": false, + "temperature": true, + "release_date": "2025-05-29", + "last_updated": "2025-06", "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 163000, - "input": 163000, - "output": 65536 - }, - "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "context": 512, + "output": 0 } }, - "Doctor-Shotgun/MS3.2-24B-Magnum-Diamond": { - "id": "Doctor-Shotgun/MS3.2-24B-Magnum-Diamond", - "name": "MS3.2 24B Magnum Diamond", - "family": "mistral", + "bfl/flux-2-flex": { + "id": "bfl/flux-2-flex", + "name": "FLUX.2 [flex]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "temperature": true, + "release_date": "2025-11-25", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 32768 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 0, + "output": 0 } }, - "NeverSleep/Llama-3-Lumimaid-70B-v0.1": { - "id": "NeverSleep/Llama-3-Lumimaid-70B-v0.1", - "name": "Lumimaid 70b", - "family": "llama", - "attachment": true, + "bfl/flux-pro-1.1-ultra": { + "id": "bfl/flux-pro-1.1-ultra", + "name": "FLUX1.1 [pro] Ultra", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "temperature": true, + "release_date": "2024-11-01", + "last_updated": "2024-11", "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 - }, - "cost": { - "input": 2.006, - "output": 2.006 + "context": 512, + "output": 0 } }, - "NeverSleep/Lumimaid-v0.2-70B": { - "id": "NeverSleep/Lumimaid-v0.2-70B", - "name": "Lumimaid v0.2", - "family": "llama", + "bfl/flux-2-max": { + "id": "bfl/flux-2-max", + "name": "FLUX.2 [max]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 - }, - "cost": { - "input": 1, - "output": 1.5 + "context": 67300, + "output": 67300 } }, - "Steelskull/L3.3-Cu-Mai-R1-70b": { - "id": "Steelskull/L3.3-Cu-Mai-R1-70b", - "name": "Llama 3.3 70B Cu Mai", - "family": "llama", + "bfl/flux-pro-1.1": { + "id": "bfl/flux-pro-1.1", + "name": "FLUX1.1 [pro]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2024-10-02", + "last_updated": "2024-10", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 512, + "output": 0 } }, - "Steelskull/L3.3-Nevoria-R1-70b": { - "id": "Steelskull/L3.3-Nevoria-R1-70b", - "name": "Steelskull Nevoria R1 70b", - "family": "llama", + "bfl/flux-pro-1.0-fill": { + "id": "bfl/flux-pro-1.0-fill", + "name": "FLUX.1 Fill [pro]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2024-10-01", + "last_updated": "2024-10", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 512, + "output": 0 } }, - "Steelskull/L3.3-MS-Evayale-70B": { - "id": "Steelskull/L3.3-MS-Evayale-70B", - "name": "Evayale 70b ", - "family": "llama", + "bfl/flux-2-klein-4b": { + "id": "bfl/flux-2-klein-4b", + "name": "FLUX.2 [klein] 4B", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 0, + "output": 0 } }, - "Steelskull/L3.3-Electra-R1-70b": { - "id": "Steelskull/L3.3-Electra-R1-70b", - "name": "Steelskull Electra R1 70b", - "family": "llama", + "bfl/flux-2-klein-9b": { + "id": "bfl/flux-2-klein-9b", + "name": "FLUX.2 [klein] 9B", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.69989, - "output": 0.69989 + "context": 0, + "output": 0 } }, - "Steelskull/L3.3-MS-Nevoria-70b": { - "id": "Steelskull/L3.3-MS-Nevoria-70b", - "name": "Steelskull Nevoria 70b", - "family": "llama", + "bfl/flux-kontext-pro": { + "id": "bfl/flux-kontext-pro", + "name": "FLUX.1 Kontext Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-05-29", + "last_updated": "2025-06", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 512, + "output": 0 } }, - "Steelskull/L3.3-MS-Evalebis-70b": { - "id": "Steelskull/L3.3-MS-Evalebis-70b", - "name": "MS Evalebis 70b", - "family": "llama", + "bfl/flux-2-pro": { + "id": "bfl/flux-2-pro", + "name": "FLUX.2 [pro]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2025-11-25", + "last_updated": "2026-06-08", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "context": 67300, + "output": 67300 } }, - "miromind-ai/mirothinker-v1.5-235b": { - "id": "miromind-ai/mirothinker-v1.5-235b", - "name": "MiroThinker v1.5 235B", - "family": "gpt", + "recraft/recraft-v4.1-pro": { + "id": "recraft/recraft-v4.1-pro", + "name": "Recraft V4.1 Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-01-07", - "last_updated": "2026-01-07", + "temperature": true, + "release_date": "2026-05-14", + "last_updated": "2026-05-14", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 4000 - }, - "cost": { - "input": 0.3, - "output": 1.2 + "context": 0, + "output": 0 } }, - "pamanseau/OpenReasoning-Nemotron-32B": { - "id": "pamanseau/OpenReasoning-Nemotron-32B", - "name": "OpenReasoning Nemotron 32B", - "family": "nemotron", + "recraft/recraft-v4.1": { + "id": "recraft/recraft-v4.1", + "name": "Recraft V4.1", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "temperature": true, + "release_date": "2026-05-14", + "last_updated": "2026-05-14", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 65536 - }, - "cost": { - "input": 0.1, - "output": 0.4 + "context": 0, + "output": 0 } }, - "arcee-ai/trinity-mini": { - "id": "arcee-ai/trinity-mini", - "name": "Trinity Mini", - "family": "trinity-mini", + "recraft/recraft-v4": { + "id": "recraft/recraft-v4", + "name": "Recraft V4", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "temperature": true, + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 - }, - "cost": { - "input": 0.045000000000000005, - "output": 0.15 + "context": 0, + "output": 0 } }, - "arcee-ai/trinity-large": { - "id": "arcee-ai/trinity-large", - "name": "Trinity Large", - "family": "trinity", + "recraft/recraft-v4-pro": { + "id": "recraft/recraft-v4-pro", + "name": "Recraft V4 Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "temperature": true, + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 - }, - "cost": { - "input": 0.25, - "output": 1 + "context": 0, + "output": 0 } }, - "cognitivecomputations/dolphin-2.9.2-qwen2-72b": { - "id": "cognitivecomputations/dolphin-2.9.2-qwen2-72b", - "name": "Dolphin 72b", - "family": "qwen", + "recraft/recraft-v2": { + "id": "recraft/recraft-v2", + "name": "Recraft V2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "temperature": true, + "release_date": "2024-03-13", + "last_updated": "2024-03", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 8192, - "output": 4096 - }, - "cost": { - "input": 0.306, - "output": 0.306 + "context": 512, + "output": 0 } }, - "deepcogito/cogito-v1-preview-qwen-32B": { - "id": "deepcogito/cogito-v1-preview-qwen-32B", - "name": "Cogito v1 Preview Qwen 32B", - "family": "qwen", + "recraft/recraft-v3": { + "id": "recraft/recraft-v3", + "name": "Recraft V3", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-05-10", - "last_updated": "2025-05-10", + "temperature": true, + "release_date": "2024-10-30", + "last_updated": "2024-10", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 - }, - "cost": { - "input": 1.7999999999999998, - "output": 1.7999999999999998 + "context": 512, + "output": 0 } }, - "deepcogito/cogito-v2.1-671b": { - "id": "deepcogito/cogito-v2.1-671b", - "name": "Cogito v2.1 671B MoE", - "family": "cogito", + "recraft/recraft-v4.1-utility-pro": { + "id": "recraft/recraft-v4.1-utility-pro", + "name": "Recraft V4.1 Utility Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "temperature": true, + "release_date": "2026-05-14", + "last_updated": "2026-05-14", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 1.25, - "output": 1.25 + "context": 0, + "output": 0 } }, - "Salesforce/Llama-xLAM-2-70b-fc-r": { - "id": "Salesforce/Llama-xLAM-2-70b-fc-r", - "name": "Llama-xLAM-2 70B fc-r", - "family": "llama", + "recraft/recraft-v4.1-utility": { + "id": "recraft/recraft-v4.1-utility", + "name": "Recraft V4.1 Utility", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "recraft", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-13", - "last_updated": "2025-04-13", + "temperature": true, + "release_date": "2026-05-14", + "last_updated": "2026-05-14", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 - }, - "cost": { - "input": 2.5, - "output": 2.5 + "context": 0, + "output": 0 } }, - "NousResearch 2/hermes-4-405b:thinking": { - "id": "NousResearch 2/hermes-4-405b:thinking", - "name": "Hermes 4 Large (Thinking)", - "family": "nousresearch", + "arcee-ai/trinity-large-preview": { + "id": "arcee-ai/trinity-large-preview", + "name": "Trinity Large Preview", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "trinity", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2026-01-27", + "last_updated": "2025-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 131000, + "output": 131000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.25, + "output": 1 } }, - "NousResearch 2/DeepHermes-3-Mistral-24B-Preview": { - "id": "NousResearch 2/DeepHermes-3-Mistral-24B-Preview", - "name": "DeepHermes-3 Mistral 24B (Preview)", - "family": "nousresearch", + "arcee-ai/trinity-large-thinking": { + "id": "arcee-ai/trinity-large-thinking", + "name": "Trinity Large Thinking", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "trinity", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-10", - "last_updated": "2025-05-10", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 262100, + "output": 80000 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.25, + "output": 0.8999999999999999 } }, - "NousResearch 2/Hermes-4-70B:thinking": { - "id": "NousResearch 2/Hermes-4-70B:thinking", - "name": "Hermes 4 (Thinking)", - "family": "nousresearch", + "arcee-ai/trinity-mini": { + "id": "arcee-ai/trinity-mini", + "name": "Trinity Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "trinity", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-09-17", - "last_updated": "2025-09-17", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-01", + "last_updated": "2025-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.2006, - "output": 0.39949999999999997 + "input": 0.045, + "output": 0.15 } }, - "NousResearch 2/hermes-4-405b": { - "id": "NousResearch 2/hermes-4-405b", - "name": "Hermes 4 Large", - "family": "nousresearch", + "perplexity/sonar-reasoning-pro": { + "id": "perplexity/sonar-reasoning-pro", + "name": "Sonar Reasoning Pro", + "description": "Web-grounded reasoning model for multi-step research and cited answers", + "family": "sonar-reasoning", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 - }, - "cost": { - "input": 0.3, - "output": 1.2 + "context": 127000, + "output": 8000 } }, - "NousResearch 2/hermes-3-llama-3.1-70b": { - "id": "NousResearch 2/hermes-3-llama-3.1-70b", - "name": "Hermes 3 70B", - "family": "nousresearch", - "attachment": false, + "perplexity/sonar": { + "id": "perplexity/sonar", + "name": "Sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-07", - "last_updated": "2026-01-07", + "tool_call": true, + "temperature": true, + "knowledge": "2025-02", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 - }, - "cost": { - "input": 0.408, - "output": 0.408 + "context": 127000, + "output": 8000 } }, - "NousResearch 2/hermes-4-70b": { - "id": "NousResearch 2/hermes-4-70b", - "name": "Hermes 4 Medium", - "family": "nousresearch", - "attachment": false, + "perplexity/sonar-pro": { + "id": "perplexity/sonar-pro", + "name": "Sonar Pro", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "family": "sonar-pro", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-03", - "last_updated": "2025-07-03", + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 - }, - "cost": { - "input": 0.2006, - "output": 0.39949999999999997 + "context": 200000, + "output": 8000 } }, - "soob3123/Veiled-Calla-12B": { - "id": "soob3123/Veiled-Calla-12B", - "name": "Veiled Calla 12B", - "family": "llama", + "amazon/titan-embed-text-v2": { + "id": "amazon/titan-embed-text-v2", + "name": "Titan Text Embeddings V2", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "titan-embed", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-13", - "last_updated": "2025-04-13", + "temperature": true, + "release_date": "2024-04-30", + "last_updated": "2024-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 - }, - "cost": { - "input": 0.3, - "output": 0.3 + "context": 8192, + "output": 1536 } }, - "soob3123/GrayLine-Qwen3-8B": { - "id": "soob3123/GrayLine-Qwen3-8B", - "name": "Grayline Qwen3 8B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "amazon/nova-2-lite": { + "id": "amazon/nova-2-lite", + "name": "Nova 2 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "nova", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-02", + "last_updated": "2024-12-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 32768 + "context": 1000000, + "output": 1000000 }, "cost": { "input": 0.3, - "output": 0.3 + "output": 2.5, + "cache_read": 0.075 } }, - "soob3123/amoral-gemma3-27B-v2": { - "id": "soob3123/amoral-gemma3-27B-v2", - "name": "Amoral Gemma3 27B v2", - "family": "gemma", - "attachment": false, + "amazon/nova-lite": { + "id": "amazon/nova-lite", + "name": "Nova Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-lite", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-23", - "last_updated": "2025-05-23", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, + "context": 300000, "output": 8192 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.06, + "output": 0.24, + "cache_read": 0.015 } }, - "nex-agi/deepseek-v3.1-nex-n1": { - "id": "nex-agi/deepseek-v3.1-nex-n1", - "name": "DeepSeek V3.1 Nex N1", - "family": "deepseek", + "amazon/nova-micro": { + "id": "amazon/nova-micro", + "name": "Nova Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-micro", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-10", - "last_updated": "2025-12-10", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { "input": ["text"], "output": ["text"] @@ -11534,999 +13334,1122 @@ "open_weights": false, "limit": { "context": 128000, - "input": 128000, "output": 8192 }, "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "input": 0.035, + "output": 0.14, + "cache_read": 0.00875 } }, - "Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B": { - "id": "Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B", - "name": "Llama 3.05 Storybreaker Ministral 70b", - "family": "llama", - "attachment": false, + "amazon/nova-pro": { + "id": "amazon/nova-pro", + "name": "Nova Pro", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova-pro", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, + "context": 300000, "output": 8192 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.8, + "output": 3.2, + "cache_read": 0.2 } }, - "Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B": { - "id": "Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B", - "name": "Nemotron Tenyxchat Storybreaker 70b", - "family": "nemotron", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "alibaba/qwen3-vl-thinking": { + "id": "alibaba/qwen3-vl-thinking", + "name": "Qwen3 VL Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-23", + "last_updated": "2025-09-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.4, + "output": 4 } }, - "anthracite-org/magnum-v4-72b": { - "id": "anthracite-org/magnum-v4-72b", - "name": "Magnum v4 72B", - "family": "llama", - "attachment": true, + "alibaba/qwen3-coder-plus": { + "id": "alibaba/qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Hosted Qwen coder for software agents, repo edits, and long-context code", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 2.006, - "output": 2.992 + "input": 1, + "output": 5, + "cache_read": 0.2 } }, - "anthracite-org/magnum-v2-72b": { - "id": "anthracite-org/magnum-v2-72b", - "name": "Magnum V2 72B", - "family": "llama", + "alibaba/wan-v2.6-r2v": { + "id": "alibaba/wan-v2.6-r2v", + "name": "Wan v2.6 Reference-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 - }, - "cost": { - "input": 2.006, - "output": 2.992 + "context": 0, + "output": 0 } }, - "ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0": { - "id": "ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0", - "name": "Omega Directive 24B Unslop v2.0", - "family": "llama", + "alibaba/qwen3-embedding-0.6b": { + "id": "alibaba/qwen3-embedding-0.6b", + "name": "Qwen3 Embedding 0.6B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "temperature": true, + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, + "context": 32768, "output": 32768 - }, - "cost": { - "input": 0.5, - "output": 0.5 } }, - "ReadyArt/The-Omega-Abomination-L-70B-v1.0": { - "id": "ReadyArt/The-Omega-Abomination-L-70B-v1.0", - "name": "The Omega Abomination V1", - "family": "llama", + "alibaba/qwen3-max-preview": { + "id": "alibaba/qwen3-max-preview", + "name": "Qwen3 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-05", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.7, - "output": 0.95 + "input": 1.2, + "output": 6, + "cache_read": 0.24 } }, - "undi95/remm-slerp-l2-13b": { - "id": "undi95/remm-slerp-l2-13b", - "name": "ReMM SLERP 13B", - "family": "llama", - "attachment": true, + "alibaba/qwen3-embedding-8b": { + "id": "alibaba/qwen3-embedding-8b", + "name": "Qwen3 Embedding 8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 6144, - "input": 6144, - "output": 4096 - }, - "cost": { - "input": 0.7989999999999999, - "output": 1.2069999999999999 + "context": 32768, + "output": 32768 } }, - "MarinaraSpaghetti/NemoMix-Unleashed-12B": { - "id": "MarinaraSpaghetti/NemoMix-Unleashed-12B", - "name": "NemoMix 12B Unleashed", - "family": "mistral-nemo", + "alibaba/qwen3-next-80b-a3b-instruct": { + "id": "alibaba/qwen3-next-80b-a3b-instruct", + "name": "Qwen3 Next 80B A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-11", + "last_updated": "2025-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.15, + "output": 1.2 } }, - "allenai/molmo-2-8b": { - "id": "allenai/molmo-2-8b", - "name": "Molmo 2 8B", - "family": "allenai", + "alibaba/qwen3.7-plus": { + "id": "alibaba/qwen3.7-plus", + "name": "Qwen 3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen3.7-plus", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 36864, - "input": 36864, - "output": 36864 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.4, + "output": 1.6, + "cache_read": 0.08, + "cache_write": 0.5 } }, - "allenai/olmo-3.1-32b-instruct": { - "id": "allenai/olmo-3.1-32b-instruct", - "name": "Olmo 3.1 32B Instruct", - "family": "allenai", + "alibaba/wan-v2.7-r2v": { + "id": "alibaba/wan-v2.7-r2v", + "name": "Wan v2.7 Reference-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-01-25", - "last_updated": "2026-01-25", + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 - }, - "cost": { - "input": 0.2, - "output": 0.6 + "context": 0, + "output": 0 } }, - "allenai/olmo-3.1-32b-think": { - "id": "allenai/olmo-3.1-32b-think", - "name": "Olmo 3.1 32B Think", - "family": "allenai", - "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-25", - "last_updated": "2026-01-25", + "alibaba/qwen3-vl-instruct": { + "id": "alibaba/qwen3-vl-instruct", + "name": "Qwen3 VL Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 + "context": 131072, + "output": 129024 }, "cost": { - "input": 0.15, - "output": 0.5 + "input": 0.4, + "output": 1.6 } }, - "allenai/olmo-3-32b-think": { - "id": "allenai/olmo-3-32b-think", - "name": "Olmo 3 32B Think", - "family": "allenai", - "attachment": false, + "alibaba/qwen3.7-max": { + "id": "alibaba/qwen3.7-max", + "name": "Qwen 3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 991000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 0.44999999999999996 + "input": 1.25, + "output": 3.75, + "cache_read": 0.25, + "cache_write": 1.5625 } }, - "stepfun-ai/step-3.5-flash:thinking": { - "id": "stepfun-ai/step-3.5-flash:thinking", - "name": "Step 3.5 Flash Thinking", - "family": "step", + "alibaba/wan-v2.5-t2v-preview": { + "id": "alibaba/wan-v2.5-t2v-preview", + "name": "Wan v2.5 Text-to-Video Preview", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "temperature": true, + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 256000 - }, - "cost": { - "input": 0.2, - "output": 0.5 + "context": 0, + "output": 0 } }, - "stepfun-ai/step-3.5-flash": { - "id": "stepfun-ai/step-3.5-flash", - "name": "Step 3.5 Flash", - "family": "step", + "alibaba/qwen3-max": { + "id": "alibaba/qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 256000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 1.2, + "output": 6, + "cache_read": 0.24 } }, - "zai-org/glm-4.7": { - "id": "zai-org/glm-4.7", - "name": "GLM 4.7", - "family": "glm", + "alibaba/qwen3-next-80b-a3b-thinking": { + "id": "alibaba/qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B Thinking", + "description": "Efficient Qwen thinking model for local reasoning, math, and coding agents", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-11", + "last_updated": "2025-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { "input": 0.15, - "output": 0.8 + "output": 1.2 } }, - "zai-org/glm-5": { - "id": "zai-org/glm-5", - "name": "GLM 5", - "family": "glm", + "alibaba/qwen3-embedding-4b": { + "id": "alibaba/qwen3-embedding-4b", + "name": "Qwen3 Embedding 4B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 128000 - }, - "cost": { - "input": 0.3, - "output": 2.55 + "context": 32768, + "output": 32768 } }, - "zai-org/glm-5.1": { - "id": "zai-org/glm-5.1", - "name": "GLM 5.1", - "family": "glm", - "attachment": false, + "alibaba/qwen3.5-flash": { + "id": "alibaba/qwen3.5-flash", + "name": "Qwen 3.5 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "temperature": true, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 131072 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 2.55 + "input": 0.1, + "output": 0.4, + "cache_read": 0.001, + "cache_write": 0.125 } }, - "zai-org/glm-5.1:thinking": { - "id": "zai-org/glm-5.1:thinking", - "name": "GLM 5.1 Thinking", - "family": "glm", + "alibaba/qwen3-coder": { + "id": "alibaba/qwen3-coder", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-22", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.55 + "input": 1.5, + "output": 7.5, + "cache_read": 0.3 } }, - "zai-org/glm-5:thinking": { - "id": "zai-org/glm-5:thinking", - "name": "GLM 5 Thinking", - "family": "glm", + "alibaba/qwen-3-235b": { + "id": "alibaba/qwen-3-235b", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 128000 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 2.55 + "input": 0.22, + "output": 0.88 } }, - "zai-org/glm-4.7-flash": { - "id": "zai-org/glm-4.7-flash", - "name": "GLM 4.7 Flash", - "family": "glm-flash", - "attachment": false, + "alibaba/qwen3.5-plus": { + "id": "alibaba/qwen3.5-plus", + "name": "Qwen 3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "input": 200000, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.07, - "output": 0.4 + "input": 0.4, + "output": 2.4, + "cache_read": 0.04, + "cache_write": 0.5 } }, - "featherless-ai/Qwerky-72B": { - "id": "featherless-ai/Qwerky-72B", - "name": "Qwerky 72B", - "family": "qwerky", + "alibaba/wan-v2.6-t2v": { + "id": "alibaba/wan-v2.6-t2v", + "name": "Wan v2.6 Text-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-20", - "last_updated": "2025-03-20", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 8192 - }, - "cost": { - "input": 0.5, - "output": 0.5 + "context": 0, + "output": 0 } }, - "mlabonne/NeuralDaredevil-8B-abliterated": { - "id": "mlabonne/NeuralDaredevil-8B-abliterated", - "name": "Neural Daredevil 8B abliterated", - "family": "llama", + "alibaba/qwen3-max-thinking": { + "id": "alibaba/qwen3-max-thinking", + "name": "Qwen 3 Max Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-23", + "last_updated": "2025-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.44, - "output": 0.44 + "input": 1.2, + "output": 6, + "cache_read": 0.24 } }, - "raifle/sorcererlm-8x22b": { - "id": "raifle/sorcererlm-8x22b", - "name": "SorcererLM 8x22B", - "family": "mixtral", - "attachment": true, + "alibaba/wan-v2.6-i2v-flash": { + "id": "alibaba/wan-v2.6-i2v-flash", + "name": "Wan v2.6 Image-to-Video Flash", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "o", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 16000, - "input": 16000, - "output": 8192 - }, - "cost": { - "input": 4.505, - "output": 4.505 + "context": 0, + "output": 0 } }, - "mistralai/mixtral-8x7b-instruct-v0.1": { - "id": "mistralai/mixtral-8x7b-instruct-v0.1", - "name": "Mixtral 8x7B", - "family": "mixtral", + "alibaba/wan-v2.6-r2v-flash": { + "id": "alibaba/wan-v2.6-r2v-flash", + "name": "Wan v2.6 Reference-to-Video Flash", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 - }, - "cost": { - "input": 0.27, - "output": 0.27 + "context": 0, + "output": 0 } }, - "mistralai/mistral-saba": { - "id": "mistralai/mistral-saba", - "name": "Mistral Saba", - "family": "mistral", + "alibaba/qwen3-coder-next": { + "id": "alibaba/qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-07-22", + "last_updated": "2026-02-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 32768 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.1989, - "output": 0.595 + "input": 0.5, + "output": 1.2 } }, - "mistralai/mistral-large-3-675b-instruct-2512": { - "id": "mistralai/mistral-large-3-675b-instruct-2512", - "name": "Mistral Large 3 675B", - "family": "mistral-large", + "alibaba/qwen3.6-27b": { + "id": "alibaba/qwen3.6-27b", + "name": "Qwen 3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 131072 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 262144, + "context": 256000, "output": 256000 }, "cost": { - "input": 1, - "output": 3 + "input": 0.6, + "output": 3.6 } }, - "mistralai/devstral-2-123b-instruct-2512": { - "id": "mistralai/devstral-2-123b-instruct-2512", - "name": "Devstral 2 123B", - "family": "devstral", + "alibaba/wan-v2.6-i2v": { + "id": "alibaba/wan-v2.6-i2v", + "name": "Wan v2.6 Image-to-Video", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "temperature": true, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 262144, - "output": 65536 - }, - "cost": { - "input": 0.4, - "output": 1.4 + "context": 0, + "output": 0 } }, - "mistralai/codestral-2508": { - "id": "mistralai/codestral-2508", - "name": "Codestral 2508", - "family": "codestral", + "alibaba/wan-v2.7-t2v": { + "id": "alibaba/wan-v2.7-t2v", + "name": "Wan v2.7 Text-to-Video", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "o", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 - }, - "cost": { - "input": 0.3, - "output": 0.8999999999999999 + "context": 0, + "output": 0 } }, - "mistralai/ministral-14b-instruct-2512": { - "id": "mistralai/ministral-14b-instruct-2512", - "name": "Ministral 3 14B", - "family": "ministral", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "alibaba/qwen-3-30b": { + "id": "alibaba/qwen-3-30b", + "name": "Qwen3-30B-A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 262144, - "output": 32768 + "context": 40960, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.12, + "output": 0.5 } }, - "mistralai/mistral-tiny": { - "id": "mistralai/mistral-tiny", - "name": "Mistral Tiny", - "family": "mistral", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2023-12-11", - "last_updated": "2024-01-01", + "alibaba/qwen3-235b-a22b-thinking": { + "id": "alibaba/qwen3-235b-a22b-thinking", + "name": "Qwen3 235B A22B Thinking 2507", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-04", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "input": 32000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.25499999999999995, - "output": 0.25499999999999995 + "input": 0.4, + "output": 4 } }, - "mistralai/ministral-8b-2512": { - "id": "mistralai/ministral-8b-2512", - "name": "Ministral 8B", - "family": "ministral", - "attachment": false, + "alibaba/qwen3-vl-235b-a22b-instruct": { + "id": "alibaba/qwen3-vl-235b-a22b-instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-12-04", - "last_updated": "2025-12-04", + "temperature": true, + "release_date": "2025-09-23", + "last_updated": "2026-05-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 262144, - "output": 32768 + "context": 131072, + "output": 129024 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.4, + "output": 1.6 } }, - "mistralai/mixtral-8x22b-instruct-v0.1": { - "id": "mistralai/mixtral-8x22b-instruct-v0.1", - "name": "Mixtral 8x22B", - "family": "mixtral", + "alibaba/qwen-3-14b": { + "id": "alibaba/qwen-3-14b", + "name": "Qwen3-14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 32768 + "context": 40960, + "output": 16384 }, "cost": { - "input": 0.8999999999999999, - "output": 0.8999999999999999 + "input": 0.12, + "output": 0.24 } }, - "mistralai/mistral-medium-3.1": { - "id": "mistralai/mistral-medium-3.1", - "name": "Mistral Medium 3.1", - "family": "mistral-medium", + "alibaba/qwen-3-32b": { + "id": "alibaba/qwen-3-32b", + "name": "Qwen 3.32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 38912 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.16, + "output": 0.64 } }, - "mistralai/ministral-3b-2512": { - "id": "mistralai/ministral-3b-2512", - "name": "Ministral 3B", - "family": "ministral", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-04", - "last_updated": "2025-12-04", + "alibaba/qwen-3.6-max-preview": { + "id": "alibaba/qwen-3.6-max-preview", + "name": "Qwen 3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 131072 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 32768 + "context": 240000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.3, + "output": 7.8, + "cache_read": 0.26, + "cache_write": 1.625 } }, - "mistralai/Mistral-Nemo-Instruct-2407": { - "id": "mistralai/Mistral-Nemo-Instruct-2407", - "name": "Mistral Nemo", - "family": "mistral-nemo", + "alibaba/qwen3-coder-30b-a3b": { + "id": "alibaba/qwen3-coder-30b-a3b", + "name": "Qwen 3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-31", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, + "context": 262144, "output": 8192 }, "cost": { - "input": 0.1003, - "output": 0.1207 + "input": 0.15, + "output": 0.6 } }, - "mistralai/mistral-medium-3": { - "id": "mistralai/mistral-medium-3", - "name": "Mistral Medium 3", - "family": "mistral-medium", + "alibaba/qwen3.6-plus": { + "id": "alibaba/qwen3.6-plus", + "name": "Qwen 3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 131072 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.5, + "output": 3, + "cache_read": 0.1, + "cache_write": 0.625 } }, - "mistralai/mistral-7b-instruct": { - "id": "mistralai/mistral-7b-instruct", - "name": "Mistral 7B Instruct", - "family": "mistral", - "attachment": true, - "reasoning": false, + "meituan/longcat-flash-thinking-2601": { + "id": "meituan/longcat-flash-thinking-2601", + "name": "LongCat Flash Thinking 2601", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "longcat", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2024-05-27", - "last_updated": "2024-05-27", + "temperature": true, + "release_date": "2026-01-15", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 32768, - "input": 32768, - "output": 8192 - }, - "cost": { - "input": 0.0544, - "output": 0.0544 + "output": 32768 } }, - "mistralai/Devstral-Small-2505": { - "id": "mistralai/Devstral-Small-2505", - "name": "Mistral Devstral Small 2505", - "family": "devstral", + "meituan/longcat-flash-chat": { + "id": "meituan/longcat-flash-chat", + "name": "LongCat Flash Chat", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "longcat", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-02", - "last_updated": "2025-08-02", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-01", + "last_updated": "2025-08-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 - }, - "cost": { - "input": 0.060000000000000005, - "output": 0.060000000000000005 + "context": 128000, + "output": 100000 } }, - "mistralai/mistral-small-creative": { - "id": "mistralai/mistral-small-creative", - "name": "Mistral Small Creative", - "family": "mistral-small", + "meta/muse-spark-1.1": { + "id": "meta/muse-spark-1.1", + "name": "Muse Spark 1.1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "muse", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "temperature": true, + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 1048576, + "output": 1048576 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 1.25, + "output": 4.25, + "cache_read": 0.15 } }, - "mistralai/mistral-large": { - "id": "mistralai/mistral-large", - "name": "Mistral Large 2411", - "family": "mistral-large", + "meta/llama-3.2-1b": { + "id": "meta/llama-3.2-1b", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-02-26", - "last_updated": "2024-02-26", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-18", "modalities": { "input": ["text"], "output": ["text"] @@ -12534,649 +14457,711 @@ "open_weights": false, "limit": { "context": 128000, - "input": 128000, - "output": 256000 + "output": 8192 }, "cost": { - "input": 2.006, - "output": 6.001 + "input": 0.1, + "output": 0.1 } }, - "mistralai/ministral-14b-2512": { - "id": "mistralai/ministral-14b-2512", - "name": "Ministral 14B", - "family": "ministral", - "attachment": false, + "meta/llama-3.2-11b": { + "id": "meta/llama-3.2-11b", + "name": "Llama 3.2 11B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-04", - "last_updated": "2025-12-04", + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 262144, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.16, + "output": 0.16 } }, - "shisa-ai/shisa-v2.1-llama3.3-70b": { - "id": "shisa-ai/shisa-v2.1-llama3.3-70b", - "name": "Shisa V2.1 Llama 3.3 70B", + "meta/llama-3.1-8b": { + "id": "meta/llama-3.1-8b", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 4096 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 0.5 + "input": 0.22, + "output": 0.22 } }, - "shisa-ai/shisa-v2-llama3.3-70b": { - "id": "shisa-ai/shisa-v2-llama3.3-70b", - "name": "Shisa V2 Llama 3.3 70B", + "meta/llama-3.2-90b": { + "id": "meta/llama-3.2-90b", + "name": "Llama 3.2 90B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", "family": "llama", - "attachment": false, + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "input": 128000, - "output": 16384 + "output": 8192 }, "cost": { - "input": 0.5, - "output": 0.5 + "input": 0.72, + "output": 0.72 } }, - "meta-llama/llama-3.3-70b-instruct": { - "id": "meta-llama/llama-3.3-70b-instruct", - "name": "Llama 3.3 70b Instruct", + "meta/llama-3.1-70b": { + "id": "meta/llama-3.1-70b", + "name": "Llama 3.1 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.23 + "input": 0.72, + "output": 0.72 } }, - "meta-llama/llama-4-scout": { - "id": "meta-llama/llama-4-scout", - "name": "Llama 4 Scout", + "meta/llama-3.2-3b": { + "id": "meta/llama-3.2-3b", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", - "attachment": true, + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "tool_call": false, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 328000, - "input": 328000, - "output": 65536 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.085, - "output": 0.46 + "input": 0.15, + "output": 0.15 } }, - "meta-llama/llama-4-maverick": { - "id": "meta-llama/llama-4-maverick", - "name": "Llama 4 Maverick", + "meta/llama-4-scout": { + "id": "meta/llama-4-scout", + "name": "Llama-4-Scout-17B-16E-Instruct-FP8", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", "family": "llama", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "input": 1048576, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.18000000000000002, - "output": 0.8 + "input": 0, + "output": 0 } }, - "meta-llama/llama-3.2-90b-vision-instruct": { - "id": "meta-llama/llama-3.2-90b-vision-instruct", - "name": "Llama 3.2 Medium", + "meta/llama-3.3-70b": { + "id": "meta/llama-3.3-70b", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", - "attachment": false, + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.9009999999999999, - "output": 0.9009999999999999 + "input": 0, + "output": 0 } }, - "meta-llama/llama-3.2-3b-instruct": { - "id": "meta-llama/llama-3.2-3b-instruct", - "name": "Llama 3.2 3b Instruct", + "meta/llama-4-maverick": { + "id": "meta/llama-4-maverick", + "name": "Llama-4-Maverick-17B-128E-Instruct-FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "tool_call": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 8192 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.0306, - "output": 0.0493 + "input": 0, + "output": 0 } }, - "meta-llama/llama-3.1-8b-instruct": { - "id": "meta-llama/llama-3.1-8b-instruct", - "name": "Llama 3.1 8b Instruct", - "family": "llama", + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-23", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.0544, - "output": 0.0544 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "GalrionSoftworks/MN-LooseCannon-12B-v1": { - "id": "GalrionSoftworks/MN-LooseCannon-12B-v1", - "name": "MN-LooseCannon-12B-v1", - "family": "mistral-nemo", + "deepseek/deepseek-v3.1-terminus": { + "id": "deepseek/deepseek-v3.1-terminus", + "name": "DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.27, + "output": 1, + "cache_read": 0.135 } }, - "baseten/Kimi-K2-Instruct-FP4": { - "id": "baseten/Kimi-K2-Instruct-FP4", - "name": "Kimi K2 0711 Instruct FP4", - "family": "kimi", + "deepseek/deepseek-v3": { + "id": "deepseek/deepseek-v3", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-12-26", + "last_updated": "2024-12-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 131072 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.1, - "output": 2 + "input": 0.27, + "output": 1.12, + "cache_read": 0.135 } }, - "Gryphe/MythoMax-L2-13b": { - "id": "Gryphe/MythoMax-L2-13b", - "name": "MythoMax 13B", - "family": "llama", + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 4000, - "input": 4000, - "output": 4096 - }, - "cost": { - "input": 0.1003, - "output": 0.1003 - } - }, - "x-ai/grok-4-fast:thinking": { - "id": "x-ai/grok-4-fast:thinking", - "name": "Grok 4 Fast Thinking", - "family": "grok", - "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-23", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "input": 2000000, - "output": 131072 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } }, - "x-ai/grok-4-07-09": { - "id": "x-ai/grok-4-07-09", - "name": "Grok 4", - "family": "grok", - "attachment": true, + "deepseek/deepseek-v3.2-thinking": { + "id": "deepseek/deepseek-v3.2-thinking", + "name": "DeepSeek V3.2 Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "reasoning_options": [], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 131072 + "context": 128000, + "output": 8000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.62, + "output": 1.85 } }, - "x-ai/grok-4-fast": { - "id": "x-ai/grok-4-fast", - "name": "Grok 4 Fast", - "family": "grok", - "attachment": true, + "deepseek/deepseek-v3.1": { + "id": "deepseek/deepseek-v3.1", + "name": "DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-09-20", - "last_updated": "2025-09-20", + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "input": 2000000, - "output": 131072 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.6, + "output": 1.7 } }, - "x-ai/grok-code-fast-1": { - "id": "x-ai/grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 131072 + "context": 128000, + "output": 8000 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 0.28, + "output": 0.42, + "cache_read": 0.028 } }, - "x-ai/grok-4.1-fast": { - "id": "x-ai/grok-4.1-fast", - "name": "Grok 4.1 Fast", - "family": "grok", - "attachment": true, + "deepseek/deepseek-r1": { + "id": "deepseek/deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "input": 2000000, - "output": 131072 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 1.35, + "output": 5.4 } }, - "x-ai/grok-4.1-fast-reasoning": { - "id": "x-ai/grok-4.1-fast-reasoning", - "name": "Grok 4.1 Fast Reasoning", - "family": "grok", + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax M2.7 High Speed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "input": 2000000, - "output": 131072 + "context": 204800, + "output": 131100 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "tencent/Hunyuan-MT-7B": { - "id": "tencent/Hunyuan-MT-7B", - "name": "Hunyuan MT 7B", - "family": "hunyuan", + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 + "context": 204800, + "output": 131000 }, "cost": { - "input": 10, - "output": 20 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "microsoft/wizardlm-2-8x22b": { - "id": "microsoft/wizardlm-2-8x22b", - "name": "WizardLM-2 8x22B", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax M2.1", + "description": "Earlier MiniMax agent model for practical coding and productivity tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "microsoft/MAI-DS-R1-FP8": { - "id": "microsoft/MAI-DS-R1-FP8", - "name": "Microsoft DeepSeek R1", - "family": "deepseek", + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax-m3", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-31", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 1000000, + "output": 1000000 }, "cost": { "input": 0.3, - "output": 0.3 + "output": 1.2, + "cache_read": 0.06 } }, - "cohere/command-r": { - "id": "cohere/command-r", - "name": "Cohere: Command R", - "family": "command-r", + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-03-11", - "last_updated": "2024-03-11", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 205000, + "output": 205000 }, "cost": { - "input": 0.476, - "output": 1.428 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "cohere/command-r-plus-08-2024": { - "id": "cohere/command-r-plus-08-2024", - "name": "Cohere: Command R+", - "family": "command-r", - "attachment": false, - "reasoning": false, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "Minimax M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 204800, + "output": 131000 }, "cost": { - "input": 2.856, - "output": 14.246 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "chutesai/Mistral-Small-3.2-24B-Instruct-2506": { - "id": "chutesai/Mistral-Small-3.2-24B-Instruct-2506", - "name": "Mistral Small 3.2 24b Instruct", - "family": "chutesai", + "minimax/minimax-m2.5-highspeed": { + "id": "minimax/minimax-m2.5-highspeed", + "name": "MiniMax M2.5 High Speed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 131072 + "context": 204800, + "output": 131000 }, "cost": { - "input": 0.2, - "output": 0.4 - } - }, - "nvidia/Llama-3.1-Nemotron-Ultra-253B-v1": { - "id": "nvidia/Llama-3.1-Nemotron-Ultra-253B-v1", - "name": "Nvidia Nemotron Ultra 253B", - "family": "nemotron", + "input": 0.6, + "output": 2.4, + "cache_read": 0.03, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.1-lightning": { + "id": "minimax/minimax-m2.1-lightning", + "name": "MiniMax M2.1 Lightning", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-03", - "last_updated": "2025-07-03", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-23", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 0.8 + "input": 0.3, + "output": 2.4, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "nvidia/nemotron-3-nano-30b-a3b": { - "id": "nvidia/nemotron-3-nano-30b-a3b", - "name": "Nvidia Nemotron 3 Nano 30B", - "family": "nemotron", + "kwaipilot/kat-coder-air-v2.5": { + "id": "kwaipilot/kat-coder-air-v2.5", + "name": "Kat Coder Air V2.5", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "kat-coder", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-15", - "last_updated": "2025-12-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-07-10", + "last_updated": "2026-07-10", "modalities": { "input": ["text"], "output": ["text"] @@ -13184,9126 +15169,10421 @@ "open_weights": false, "limit": { "context": 256000, - "input": 256000, - "output": 262144 + "output": 80000 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.15, + "output": 0.6, + "cache_read": 0.03 } }, - "nvidia/nvidia-nemotron-nano-9b-v2": { - "id": "nvidia/nvidia-nemotron-nano-9b-v2", - "name": "Nvidia Nemotron Nano 9B v2", - "family": "nemotron", + "kwaipilot/kat-coder-pro-v2.5": { + "id": "kwaipilot/kat-coder-pro-v2.5", + "name": "Kat Coder Pro V2.5", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "kat-coder", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-18", - "last_updated": "2025-08-18", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-07-10", + "last_updated": "2026-07-10", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 256000, + "output": 80000 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.74, + "output": 2.96, + "cache_read": 0.15 } }, - "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF": { - "id": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", - "name": "Nvidia Nemotron 70b", - "family": "nemotron", + "kwaipilot/kat-coder-pro-v1": { + "id": "kwaipilot/kat-coder-pro-v1", + "name": "KAT-Coder-Pro V1", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "kat-coder", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-11-09", + "last_updated": "2025-10-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 256000, + "output": 32000 }, "cost": { - "input": 0.357, - "output": 0.408 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "nvidia/Llama-3.3-Nemotron-Super-49B-v1": { - "id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1", - "name": "Nvidia Nemotron Super 49B", - "family": "nemotron", + "kwaipilot/kat-coder-pro-v2": { + "id": "kwaipilot/kat-coder-pro-v2", + "name": "Kat Coder Pro V2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "kat-coder", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } - }, - "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5": { - "id": "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5", - "name": "Nvidia Nemotron Super 49B v1.5", - "family": "nemotron", + } + } + }, + "submodel": { + "id": "submodel", + "env": ["SUBMODEL_INSTAGEN_ACCESS_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://llm.submodel.ai/v1", + "name": "submodel", + "doc": "https://submodel.gitbook.io", + "models": { + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen3 235B A22B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.05, - "output": 0.25 + "input": 0.2, + "output": 0.6 } }, - "TheDrummer 2/Anubis-70B-v1": { - "id": "TheDrummer 2/Anubis-70B-v1", - "name": "Anubis 70B v1", - "family": "llama", + "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "input": 65536, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.31, - "output": 0.31 + "input": 0.2, + "output": 0.8 } }, - "TheDrummer 2/Cydonia-24B-v4.3": { - "id": "TheDrummer 2/Cydonia-24B-v4.3", - "name": "The Drummer Cydonia 24B v4.3", - "family": "llama", + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-25", - "last_updated": "2025-12-25", + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.1003, - "output": 0.1207 + "input": 0.2, + "output": 0.3 } }, - "TheDrummer 2/Magidonia-24B-v4.3": { - "id": "TheDrummer 2/Magidonia-24B-v4.3", - "name": "The Drummer Magidonia 24B v4.3", - "family": "llama", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-25", - "last_updated": "2025-12-25", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, + "context": 131072, "output": 32768 }, "cost": { - "input": 0.1003, - "output": 0.1207 + "input": 0.1, + "output": 0.5 } }, - "TheDrummer 2/Cydonia-24B-v4": { - "id": "TheDrummer 2/Cydonia-24B-v4", - "name": "The Drummer Cydonia 24B v4", - "family": "llama", + "zai-org/GLM-4.5-FP8": { + "id": "zai-org/GLM-4.5-FP8", + "name": "GLM 4.5 FP8", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.2006, - "output": 0.2414 + "input": 0.2, + "output": 0.8 } }, - "TheDrummer 2/Anubis-70B-v1.1": { - "id": "TheDrummer 2/Anubis-70B-v1.1", - "name": "Anubis 70B v1.1", - "family": "llama", + "zai-org/GLM-4.5-Air": { + "id": "zai-org/GLM-4.5-Air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "tool_call": true, + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "input": 131072, - "output": 16384 + "output": 131072 }, "cost": { - "input": 0.31, - "output": 0.31 + "input": 0.1, + "output": 0.5 } }, - "TheDrummer 2/Rocinante-12B-v1.1": { - "id": "TheDrummer 2/Rocinante-12B-v1.1", - "name": "Rocinante 12b", - "family": "llama", + "deepseek-ai/DeepSeek-V3-0324": { + "id": "deepseek-ai/DeepSeek-V3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 75000, + "output": 163840 }, "cost": { - "input": 0.408, - "output": 0.595 + "input": 0.2, + "output": 0.8 } }, - "TheDrummer 2/Cydonia-24B-v4.1": { - "id": "TheDrummer 2/Cydonia-24B-v4.1", - "name": "The Drummer Cydonia 24B v4.1", - "family": "llama", + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 32768 - }, - "cost": { - "input": 0.1003, - "output": 0.1207 - } - }, - "TheDrummer 2/UnslopNemo-12B-v4.1": { - "id": "TheDrummer 2/UnslopNemo-12B-v4.1", - "name": "UnslopNemo 12b v4", - "family": "llama", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", - "modalities": { - "input": ["text", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 32768, - "input": 32768, - "output": 8192 + "context": 75000, + "output": 163840 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.5, + "output": 2.15 } }, - "TheDrummer 2/Cydonia-24B-v2": { - "id": "TheDrummer 2/Cydonia-24B-v2", - "name": "The Drummer Cydonia 24B v2", - "family": "llama", + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-23", + "last_updated": "2025-08-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "input": 16384, - "output": 32768 + "context": 75000, + "output": 163840 }, "cost": { - "input": 0.1003, - "output": 0.1207 + "input": 0.2, + "output": 0.8 } - }, - "TheDrummer 2/skyfall-36b-v2": { - "id": "TheDrummer 2/skyfall-36b-v2", - "name": "TheDrummer Skyfall 36B V2", + } + } + }, + "huggingface": { + "id": "huggingface", + "env": ["HF_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://router.huggingface.co/v1", + "name": "Hugging Face", + "doc": "https://huggingface.co/docs/inference-providers", + "models": { + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", "family": "llama", - "attachment": true, + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 64000, - "input": 64000, - "output": 32768 + "context": 131072, + "output": 4096 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.59, + "output": 0.79 } }, - "deepseek-ai/DeepSeek-V3.1:thinking": { - "id": "deepseek-ai/DeepSeek-V3.1:thinking", - "name": "DeepSeek V3.1 Thinking", - "family": "deepseek-thinking", + "moonshotai/Kimi-K2-Thinking": { + "id": "moonshotai/Kimi-K2-Thinking", + "name": "Kimi-K2-Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.7 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": true, + "moonshotai/Kimi-K2-Instruct-0905": { + "id": "moonshotai/Kimi-K2-Instruct-0905", + "name": "Kimi-K2-Instruct-0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-04", + "last_updated": "2025-09-04", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.7 + "input": 1, + "output": 3 } }, - "deepseek-ai/DeepSeek-V3.1-Terminus:thinking": { - "id": "deepseek-ai/DeepSeek-V3.1-Terminus:thinking", - "name": "DeepSeek V3.1 Terminus (Thinking)", - "family": "deepseek-thinking", + "moonshotai/Kimi-K2-Instruct": { + "id": "moonshotai/Kimi-K2-Instruct", + "name": "Kimi-K2-Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-07-14", + "last_updated": "2025-07-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 0.7 + "input": 1, + "output": 3 } }, - "deepseek-ai/deepseek-v3.2-exp-thinking": { - "id": "deepseek-ai/deepseek-v3.2-exp-thinking", - "name": "DeepSeek V3.2 Exp Thinking", - "family": "deepseek-thinking", - "attachment": false, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi-K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "input": 163840, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "deepseek-ai/deepseek-v3.2-exp": { - "id": "deepseek-ai/deepseek-v3.2-exp", - "name": "DeepSeek V3.2 Exp", - "family": "deepseek", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-01", + "last_updated": "2026-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "input": 163840, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27999999999999997, - "output": 0.42000000000000004 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek", - "attachment": false, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 163840 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.4, - "output": 1.7 + "input": 0.95, + "output": 4 } }, - "deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", + "stepfun-ai/Step-3.5-Flash": { + "id": "stepfun-ai/Step-3.5-Flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-08-02", - "last_updated": "2025-08-02", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-29", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 262144, + "output": 256000 }, "cost": { - "input": 0.25, - "output": 0.7 + "input": 0.1, + "output": 0.3 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "GPT 5.1 Codex Max", - "family": "gpt-codex", + "stepfun-ai/Step-3.7-Flash": { + "id": "stepfun-ai/Step-3.7-Flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 256000 }, "cost": { - "input": 2.5, - "output": 20 + "input": 0.2, + "output": 1.15 } }, - "openai/gpt-5.2-chat": { - "id": "openai/gpt-5.2-chat", - "name": "GPT 5.2 Chat", - "family": "gpt", + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2026-01-01", - "last_updated": "2026-01-01", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.14, + "output": 0.4 } }, - "openai/gpt-4o-mini-search-preview": { - "id": "openai/gpt-4o-mini-search-preview", - "name": "GPT-4o mini Search Preview", - "family": "gpt-mini", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "google/gemma-4-26B-A4B-it": { + "id": "google/gemma-4-26B-A4B-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.088, - "output": 0.35 + "input": 0.13, + "output": 0.4 } }, - "openai/chatgpt-4o-latest": { - "id": "openai/chatgpt-4o-latest", - "name": "ChatGPT 4o", - "family": "gpt", + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 4.998, - "output": 14.993999999999998 + "input": 0.15, + "output": 0.95 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT 5.2 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "Qwen/Qwen3-Coder-Next": { + "id": "Qwen/Qwen3-Coder-Next", + "name": "Qwen3-Coder-Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2026-01-01", - "last_updated": "2026-01-01", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 21, - "output": 168 + "input": 0.2, + "output": 1.5 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT 5 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "Qwen/Qwen3-Embedding-8B": { + "id": "Qwen/Qwen3-Embedding-8B", + "name": "Qwen 3 Embedding 8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": false, + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.25, - "output": 2 + "input": 0.01, + "output": 0 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT 5 Nano", - "family": "gpt-nano", + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.05, - "output": 0.4 + "input": 0.47, + "output": 3.19 } }, - "openai/gpt-4-turbo": { - "id": "openai/gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", - "attachment": true, + "Qwen/Qwen3-Next-80B-A3B-Instruct": { + "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", + "name": "Qwen3-Next-80B-A3B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2023-11-06", - "last_updated": "2024-01-01", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-11", + "last_updated": "2025-09-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 4096 + "context": 262144, + "output": 66536 }, "cost": { - "input": 10, - "output": 30 + "input": 0.25, + "output": 1 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT 5.2", - "family": "gpt", + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen3.5-397B-A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "release_date": "2026-01-01", - "last_updated": "2026-01-01", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.6, + "output": 3.6 } }, - "openai/o3-mini-high": { - "id": "openai/o3-mini-high", - "name": "OpenAI o3-mini (High)", - "family": "o-mini", + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen3-235B-A22B-Thinking-2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.64, - "output": 2.588 + "input": 0.3, + "output": 3 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", - "attachment": true, + "Qwen/Qwen3-Embedding-4B": { + "id": "Qwen/Qwen3-Embedding-4B", + "name": "Qwen 3 Embedding 4B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "temperature": false, + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 32000, + "output": 2048 }, "cost": { - "input": 0.1496, - "output": 0.595 + "input": 0.01, + "output": 0 } }, - "openai/o4-mini-deep-research": { - "id": "openai/o4-mini-deep-research", - "name": "OpenAI o4-mini Deep Research", - "family": "o-mini", + "Qwen/Qwen3-Coder-480B-A35B-Instruct": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "name": "Qwen3-Coder-480B-A35B-Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 262144, + "output": 66536 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 2, + "output": 2 } }, - "openai/gpt-5.1-chat": { - "id": "openai/gpt-5.1-chat", - "name": "GPT 5.1 Chat", - "family": "gpt", + "Qwen/Qwen3.5-122B-A10B": { + "id": "Qwen/Qwen3.5-122B-A10B", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.4, + "output": 3.2 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "OpenAI o4-mini", - "family": "o-mini", + "Qwen/Qwen3-Coder-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 0.07, + "output": 0.26 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT 5.2 Codex", - "family": "gpt-codex", + "Qwen/Qwen3.5-27B": { + "id": "Qwen/Qwen3.5-27B", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.3, + "output": 2.4 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT 5.1 Codex Mini", - "family": "gpt-codex-mini", - "attachment": true, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "Qwen/Qwen3-Next-80B-A3B-Thinking": { + "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", + "name": "Qwen3-Next-80B-A3B-Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-11", + "last_updated": "2025-09-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.25, + "input": 0.3, "output": 2 } }, - "openai/o1-preview": { - "id": "openai/o1-preview", - "name": "OpenAI o1-preview", - "family": "o", - "attachment": false, + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 14.993999999999998, - "output": 59.993 + "input": 0.17, + "output": 0.25 } }, - "openai/gpt-4o-2024-08-06": { - "id": "openai/gpt-4o-2024-08-06", - "name": "GPT-4o (2024-08-06)", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-08-06", - "last_updated": "2024-08-06", + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, + "context": 131072, "output": 16384 }, "cost": { - "input": 2.499, - "output": 9.996 + "input": 0.29, + "output": 0.59 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT 5.1", - "family": "gpt", - "attachment": true, + "Qwen/Qwen3-235B-A22B": { + "id": "Qwen/Qwen3-235B-A22B", + "name": "Qwen3 235B-A22B", + "description": "Large open Qwen MoE for multilingual reasoning, coding, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 40960, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.2, + "output": 0.8 } }, - "openai/o1": { - "id": "openai/o1", - "name": "OpenAI o1", - "family": "o", - "attachment": false, + "Qwen/Qwen3.5-35B-A3B": { + "id": "Qwen/Qwen3.5-35B-A3B", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-17", - "last_updated": "2024-12-17", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 14.993999999999998, - "output": 59.993 + "input": 0.25, + "output": 2 } }, - "openai/gpt-3.5-turbo": { - "id": "openai/gpt-3.5-turbo", - "name": "GPT-3.5 Turbo", - "family": "gpt", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2022-11-30", - "last_updated": "2024-01-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16385, - "input": 16385, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.25, + "output": 0.69 } }, - "openai/o3-deep-research": { - "id": "openai/o3-deep-research", - "name": "OpenAI o3 Deep Research", - "family": "o", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.1, + "output": 0.5 } }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "OpenAI o3-mini", - "family": "o-mini", + "XiaomiMiMo/MiMo-V2.5-Pro": { + "id": "XiaomiMiMo/MiMo-V2.5-Pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 1, + "output": 3 } }, - "openai/gpt-4-turbo-preview": { - "id": "openai/gpt-4-turbo-preview", - "name": "GPT-4 Turbo Preview", - "family": "gpt", + "XiaomiMiMo/MiMo-V2-Flash": { + "id": "XiaomiMiMo/MiMo-V2-Flash", + "name": "MiMo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2023-11-06", - "last_updated": "2024-01-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, + "context": 262144, "output": 4096 }, "cost": { - "input": 9.996, - "output": 30.004999999999995 + "input": 0.1, + "output": 0.3 } }, - "openai/o1-pro": { - "id": "openai/o1-pro", - "name": "OpenAI o1 Pro", - "family": "o-pro", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "zai-org/GLM-4.7-Flash": { + "id": "zai-org/GLM-4.7-Flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 200000, - "input": 200000, - "output": 100000 + "output": 128000 }, "cost": { - "input": 150, - "output": 600 + "input": 0, + "output": 0 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5 Codex", - "family": "gpt-codex", + "zai-org/GLM-4.6": { + "id": "zai-org/GLM-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.55, + "output": 2.2 } }, - "openai/gpt-5.1-chat-latest": { - "id": "openai/gpt-5.1-chat-latest", - "name": "GPT 5.1 Chat (Latest)", - "family": "gpt", - "attachment": true, + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 16384 + "context": 202752, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "openai/gpt-4o-search-preview": { - "id": "openai/gpt-4o-search-preview", - "name": "GPT-4o Search Preview", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.47, - "output": 5.88 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "openai/gpt-4.1-nano": { - "id": "openai/gpt-4.1-nano", - "name": "GPT 4.1 Nano", - "family": "gpt-nano", + "zai-org/GLM-4.5V": { + "id": "zai-org/GLM-4.5V", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "input": 1047576, - "output": 32768 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.6, + "output": 1.8 } }, - "openai/o4-mini-high": { - "id": "openai/o4-mini-high", - "name": "OpenAI o4-mini high", - "family": "o-mini", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 1.1, + "input": 1.4, "output": 4.4 } }, - "openai/o3": { - "id": "openai/o3", - "name": "OpenAI o3", - "family": "o", + "zai-org/GLM-4.5-Air": { + "id": "zai-org/GLM-4.5-Air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 131072, + "output": 98304 }, "cost": { - "input": 2, - "output": 8 + "input": 0.13, + "output": 0.85 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-03", + "last_updated": "2026-04-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.04, - "output": 0.15 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "GPT 5 Pro", - "family": "gpt-pro", - "attachment": true, + "zai-org/GLM-4.5": { + "id": "zai-org/GLM-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 131072, + "output": 98304 }, "cost": { - "input": 15, - "output": 120 + "input": 0.6, + "output": 2.2 } }, - "openai/gpt-5.1-2025-11-13": { - "id": "openai/gpt-5.1-2025-11-13", - "name": "GPT-5.1 (2025-11-13)", - "family": "gpt", + "deepseek-ai/DeepSeek-R1": { + "id": "deepseek-ai/DeepSeek-R1", + "name": "DeepSeek-R1", + "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, + "context": 64000, "output": 32768 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.7, + "output": 2.5 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 163840, + "output": 163840 }, "cost": { - "input": 2.499, - "output": 9.996 + "input": 3, + "output": 5 } }, - "openai/o3-mini-low": { - "id": "openai/o3-mini-low", - "name": "OpenAI o3-mini (Low)", - "family": "o-mini", + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.14, + "output": 0.28 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT 5", - "family": "gpt", - "attachment": true, + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 1048576, + "output": 393216 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "openai/gpt-oss-safeguard-20b": { - "id": "openai/gpt-oss-safeguard-20b", - "name": "GPT OSS Safeguard 20B", - "family": "gpt-oss", + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-10-29", - "last_updated": "2025-10-29", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 163840, + "output": 65536 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.28, + "output": 0.4 } }, - "openai/o3-pro-2025-06-10": { - "id": "openai/o3-pro-2025-06-10", - "name": "OpenAI o3-pro (2025-06-10)", - "family": "o-pro", + "MiniMaxAI/MiniMax-M2.1": { + "id": "MiniMaxAI/MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-06-10", - "last_updated": "2025-06-10", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-10", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 100000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 9.996, - "output": 19.992 + "input": 0.3, + "output": 1.2 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "MiniMaxAI/MiniMax-M2": { + "id": "MiniMaxAI/MiniMax-M2", + "name": "MiniMax-M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 204800, + "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.25 + "input": 0.3, + "output": 1.2 } }, - "openai/gpt-5-chat-latest": { - "id": "openai/gpt-5-chat-latest", - "name": "GPT 5 Chat", - "family": "gpt", - "attachment": true, + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT 4.1", - "family": "gpt", + "MiniMaxAI/MiniMax-M3": { + "id": "MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2025-09-10", - "last_updated": "2025-09-10", + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "input": 1047576, - "output": 32768 + "context": 524288, + "output": 128000 }, "cost": { - "input": 2, - "output": 8 + "input": 0.3, + "output": 1.2 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT 4.1 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "input": 1047576, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 1.6 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } - }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT 5.1 Codex", - "family": "gpt-codex", - "attachment": true, + } + } + }, + "minimax-coding-plan": { + "id": "minimax-coding-plan", + "env": ["MINIMAX_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.minimax.io/anthropic/v1", + "name": "MiniMax Token Plan (minimax.io)", + "doc": "https://platform.minimax.io/docs/token-plan/intro", + "models": { + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0, + "output": 0 } }, - "openai/gpt-4o-2024-11-20": { - "id": "openai/gpt-4o-2024-11-20", - "name": "GPT-4o (2024-11-20)", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-11-20", - "last_updated": "2024-11-20", + "MiniMax-M2.5-highspeed": { + "id": "MiniMax-M2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-02-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "VongolaChouko/Starcannon-Unleashed-12B-v1.0": { - "id": "VongolaChouko/Starcannon-Unleashed-12B-v1.0", - "name": "Mistral Nemo Starcannon 12b v1", - "family": "mistral-nemo", + "MiniMax-M2.7-highspeed": { + "id": "MiniMax-M2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "amazon/nova-lite-v1": { - "id": "amazon/nova-lite-v1", - "name": "Amazon Nova Lite 1.0", - "family": "nova-lite", + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 300000, - "input": 300000, - "output": 5120 + "context": 196608, + "output": 128000 }, "cost": { - "input": 0.0595, - "output": 0.238 + "input": 0, + "output": 0 } }, - "amazon/nova-pro-v1": { - "id": "amazon/nova-pro-v1", - "name": "Amazon Nova Pro 1.0", - "family": "nova-pro", + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 300000, - "input": 300000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.7989999999999999, - "output": 3.1959999999999997 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "amazon/nova-2-lite-v1": { - "id": "amazon/nova-2-lite-v1", - "name": "Amazon Nova 2 Lite", - "family": "nova", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "MiniMax-M3": { + "id": "MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 1000000, - "input": 1000000, - "output": 65535 + "output": 128000 }, "cost": { - "input": 0.5099999999999999, - "output": 4.25 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "amazon/nova-micro-v1": { - "id": "amazon/nova-micro-v1", - "name": "Amazon Nova Micro 1.0", - "family": "nova-micro", + "MiniMax-M2.7": { + "id": "MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 5120 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.0357, - "output": 0.1394 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "Sao10K/L3.3-70B-Euryale-v2.3": { - "id": "Sao10K/L3.3-70B-Euryale-v2.3", - "name": "Llama 3.3 70B Euryale", - "family": "llama", + } + } + }, + "novita-ai": { + "id": "novita-ai", + "env": ["NOVITA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.novita.ai/openai", + "name": "NovitaAI", + "doc": "https://novita.ai/docs/guides/introduction", + "models": { + "inclusionai/ling-2.6-1t": { + "id": "inclusionai/ling-2.6-1t", + "name": "Ling-2.6-1T", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "ling", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-23", + "last_updated": "2026-06-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 20480, - "input": 20480, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.3, + "output": 2.5, + "cache_read": 0.06 } }, - "Sao10K/L3.1-70B-Euryale-v2.2": { - "id": "Sao10K/L3.1-70B-Euryale-v2.2", - "name": "Llama 3.1 70B Euryale", - "family": "llama", + "inclusionai/ring-2.6-1t": { + "id": "inclusionai/ring-2.6-1t", + "name": "Ring-2.6-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "ring", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-08", + "last_updated": "2026-05-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 20480, - "input": 20480, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.306, - "output": 0.357 + "input": 0.3, + "output": 2.5, + "cache_read": 0.06 } }, - "Sao10K/L3.1-70B-Hanami-x1": { - "id": "Sao10K/L3.1-70B-Hanami-x1", - "name": "Llama 3.1 70B Hanami", - "family": "llama", + "inclusionai/ling-2.6-flash": { + "id": "inclusionai/ling-2.6-flash", + "name": "Ling-2.6-flash", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "ling", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "Sao10K/L3-8B-Stheno-v3.2": { - "id": "Sao10K/L3-8B-Stheno-v3.2", - "name": "Sao10K Stheno 8b", + "meta-llama/llama-3.1-8b-instruct": { + "id": "meta-llama/llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-11-29", - "last_updated": "2024-11-29", + "temperature": true, + "release_date": "2024-07-24", + "last_updated": "2024-07-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 16384, - "input": 16384, - "output": 8192 + "output": 16384 }, "cost": { - "input": 0.2006, - "output": 0.2006 + "input": 0.02, + "output": 0.05 } }, - "LatitudeGames/Wayfarer-Large-70B-Llama-3.3": { - "id": "LatitudeGames/Wayfarer-Large-70B-Llama-3.3", - "name": "Llama 3.3 70B Wayfarer", + "meta-llama/llama-3-70b-instruct": { + "id": "meta-llama/llama-3-70b-instruct", + "name": "Llama3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-02-20", - "last_updated": "2025-02-20", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 16384, - "input": 16384, - "output": 16384 - }, - "cost": { - "input": 0.700000007, - "output": 0.700000007 - } - }, - "z-ai/glm-4.6:thinking": { - "id": "z-ai/glm-4.6:thinking", - "name": "GLM 4.6 Thinking", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, "structured_output": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "temperature": true, + "release_date": "2024-04-25", + "last_updated": "2024-04-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 65535 + "context": 8192, + "output": 8000 }, "cost": { - "input": 0.4, - "output": 1.5 + "input": 0.51, + "output": 0.74 } }, - "z-ai/glm-4.5v": { - "id": "z-ai/glm-4.5v", - "name": "GLM 4.5V", - "family": "glmv", + "meta-llama/llama-4-scout-17b-16e-instruct": { + "id": "meta-llama/llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-22", - "last_updated": "2025-11-22", + "temperature": true, + "release_date": "2025-04-06", + "last_updated": "2025-04-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 64000, - "input": 64000, - "output": 96000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 1.7999999999999998 + "input": 0.18, + "output": 0.59 } }, - "z-ai/glm-4.6": { - "id": "z-ai/glm-4.6", - "name": "GLM 4.6", - "family": "glm", + "meta-llama/llama-3.3-70b-instruct": { + "id": "meta-llama/llama-3.3-70b-instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-07", + "last_updated": "2024-12-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 65535 + "context": 131072, + "output": 120000 }, "cost": { - "input": 0.4, - "output": 1.5 + "input": 0.135, + "output": 0.4 } }, - "z-ai/glm-4.5v:thinking": { - "id": "z-ai/glm-4.5v:thinking", - "name": "GLM 4.5V Thinking", - "family": "glmv", - "attachment": true, - "reasoning": true, + "meta-llama/llama-3-8b-instruct": { + "id": "meta-llama/llama-3-8b-instruct", + "name": "Llama 3 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-22", - "last_updated": "2025-11-22", + "temperature": true, + "release_date": "2024-04-25", + "last_updated": "2024-04-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 64000, - "input": 64000, - "output": 96000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 1.7999999999999998 + "input": 0.04, + "output": 0.04 } }, - "baidu/ernie-4.5-vl-28b-a3b": { - "id": "baidu/ernie-4.5-vl-28b-a3b", - "name": "ERNIE 4.5 VL 28B", - "family": "ernie", - "attachment": true, + "meta-llama/llama-3.2-3b-instruct": { + "id": "meta-llama/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2024-09-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 32768, - "input": 32768, - "output": 16384 + "output": 32000 }, "cost": { - "input": 0.13999999999999999, - "output": 0.5599999999999999 + "input": 0.03, + "output": 0.05 } }, - "baidu/ernie-4.5-300b-a47b": { - "id": "baidu/ernie-4.5-300b-a47b", - "name": "ERNIE 4.5 300B", - "family": "ernie", - "attachment": false, + "meta-llama/llama-4-maverick-17b-128e-instruct-fp8": { + "id": "meta-llama/llama-4-maverick-17b-128e-instruct-fp8", + "name": "Llama 4 Maverick Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "temperature": true, + "release_date": "2025-04-06", + "last_updated": "2025-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "input": 131072, - "output": 16384 + "context": 1048576, + "output": 8192 }, "cost": { - "input": 0.35, - "output": 1.15 + "input": 0.27, + "output": 0.85 } }, - "dmind/dmind-1": { - "id": "dmind/dmind-1", - "name": "DMind-1", - "family": "gpt", + "moonshotai/kimi-k2-instruct": { + "id": "moonshotai/kimi-k2-instruct", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "tool_call": true, + "temperature": true, + "release_date": "2025-07-11", + "last_updated": "2025-07-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 0.6 + "input": 0.57, + "output": 2.3 } }, - "dmind/dmind-1-mini": { - "id": "dmind/dmind-1-mini", - "name": "DMind-1-Mini", - "family": "gpt", + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2025-11-07", + "last_updated": "2026-06-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.4 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "Infermatic/MN-12B-Inferor-v0.0": { - "id": "Infermatic/MN-12B-Inferor-v0.0", - "name": "Mistral Nemo Inferor 12B", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.25499999999999995, - "output": 0.49299999999999994 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "meituan-longcat/LongCat-Flash-Chat-FP8": { - "id": "meituan-longcat/LongCat-Flash-Chat-FP8", - "name": "LongCat Flash", - "family": "longcat", - "attachment": false, - "reasoning": false, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-08-31", - "last_updated": "2025-08-31", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.15, - "output": 0.7 + "input": 0.8, + "output": 3.4, + "cache_read": 0.16 } }, - "meganova-ai/manta-mini-1.0": { - "id": "meganova-ai/manta-mini-1.0", - "name": "Manta Mini 1.0", - "family": "nova", + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-20", - "last_updated": "2025-12-20", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.02, - "output": 0.16 + "input": 0.6, + "output": 2.5 } }, - "meganova-ai/manta-pro-1.0": { - "id": "meganova-ai/manta-pro-1.0", - "name": "Manta Pro 1.0", - "family": "nova", + "minimaxai/minimax-m1-80k": { + "id": "minimaxai/minimax-m1-80k", + "name": "MiniMax M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-20", - "last_updated": "2025-12-20", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 1000000, + "output": 40000 }, "cost": { - "input": 0.060000000000000005, - "output": 0.5 + "input": 0.55, + "output": 2.2 } }, - "meganova-ai/manta-flash-1.0": { - "id": "meganova-ai/manta-flash-1.0", - "name": "Manta Flash 1.0", - "family": "nova", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-20", - "last_updated": "2025-12-20", + "baidu/ernie-4.5-vl-28b-a3b": { + "id": "baidu/ernie-4.5-vl-28b-a3b", + "name": "ERNIE 4.5 VL 28B A3B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2026-06-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 30000, + "output": 8000 }, "cost": { - "input": 0.02, - "output": 0.16 + "input": 0.14, + "output": 0.56 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "MiniMax M2.7", - "family": "minimax", + "baidu/ernie-4.5-21B-a3b": { + "id": "baidu/ernie-4.5-21B-a3b", + "name": "ERNIE 4.5 21B A3B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "ernie", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 204800, - "input": 204800, - "output": 131072 + "context": 120000, + "output": 8000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.07, + "output": 0.28 } }, - "minimax/minimax-01": { - "id": "minimax/minimax-01", - "name": "MiniMax 01", - "family": "minimax", + "baidu/ernie-4.5-vl-424b-a47b": { + "id": "baidu/ernie-4.5-vl-424b-a47b", + "name": "ERNIE 4.5 VL 424B A47B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-01-15", - "last_updated": "2025-01-15", + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000192, - "input": 1000192, - "output": 16384 + "context": 123000, + "output": 16000 }, "cost": { - "input": 0.1394, - "output": 1.1219999999999999 + "input": 0.42, + "output": 1.25 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "MiniMax M2.1", - "family": "minimax", + "baidu/ernie-4.5-21B-a3b-thinking": { + "id": "baidu/ernie-4.5-21B-a3b-thinking", + "name": "ERNIE-4.5-21B-A3B-Thinking", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "ernie", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 131072 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.33, - "output": 1.32 + "input": 0.07, + "output": 0.28 } }, - "minimax/minimax-m2-her": { - "id": "minimax/minimax-m2-her", - "name": "MiniMax M2-her", - "family": "minimax", + "baidu/ernie-4.5-300b-a47b-paddle": { + "id": "baidu/ernie-4.5-300b-a47b-paddle", + "name": "ERNIE 4.5 300B A47B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-01-24", - "last_updated": "2026-01-24", + "structured_output": true, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65532, - "input": 65532, - "output": 2048 + "context": 123000, + "output": 12000 }, "cost": { - "input": 0.30200000000000005, - "output": 1.2069999999999999 + "input": 0.28, + "output": 1.1 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "baidu/ernie-4.5-vl-28b-a3b-thinking": { + "id": "baidu/ernie-4.5-vl-28b-a3b-thinking", + "name": "ERNIE-4.5-VL-28B-A3B-Thinking", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "temperature": true, + "release_date": "2025-11-26", + "last_updated": "2025-11-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 204800, - "input": 204800, - "output": 131072 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.39, + "output": 0.39 } }, - "qwen/Qwen3.6-35B-A3B:thinking": { - "id": "qwen/Qwen3.6-35B-A3B:thinking", - "name": "Qwen3.6 35B A3B Thinking", - "family": "qwen3.6", - "attachment": false, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2026-04-19", - "last_updated": "2026-04-21", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 16384 + "output": 131072 }, "cost": { - "input": 0.29, - "output": 1.74 + "input": 0.14, + "output": 0.4 } }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 258048, - "input": 258048, - "output": 65536 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.13, + "output": 0.4 } }, - "qwen/Qwen3.6-35B-A3B": { - "id": "qwen/Qwen3.6-35B-A3B", - "name": "Qwen3.6 35B A3B", - "family": "qwen3.6", - "attachment": false, + "google/gemma-3-12b-it": { + "id": "google/gemma-3-12b-it", + "name": "Gemma 3 12B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2026-04-17", - "last_updated": "2026-04-21", + "temperature": true, + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.29, - "output": 1.74 + "input": 0.05, + "output": 0.1 } }, - "unsloth/gemma-3-1b-it": { - "id": "unsloth/gemma-3-1b-it", - "name": "Gemma 3 1B IT", - "family": "unsloth", - "attachment": false, + "google/gemma-3-27b-it": { + "id": "google/gemma-3-27b-it", + "name": "Gemma 3 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "temperature": true, + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 98304, + "output": 16384 }, "cost": { - "input": 0.1003, - "output": 0.1003 + "input": 0.119, + "output": 0.2 } }, - "unsloth/gemma-3-12b-it": { - "id": "unsloth/gemma-3-12b-it", - "name": "Gemma 3 12B IT", - "family": "unsloth", - "attachment": true, + "microsoft/wizardlm-2-8x22b": { + "id": "microsoft/wizardlm-2-8x22b", + "name": "Wizardlm 2 8x22B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "temperature": true, + "release_date": "2024-04-24", + "last_updated": "2024-04-24", "modalities": { - "input": ["text", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 131072 + "context": 65535, + "output": 8000 }, "cost": { - "input": 0.272, - "output": 0.272 + "input": 0.62, + "output": 0.62 } }, - "unsloth/gemma-3-4b-it": { - "id": "unsloth/gemma-3-4b-it", - "name": "Gemma 3 4B IT", - "family": "unsloth", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "OpenAI GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.2006, - "output": 0.2006 + "input": 0.05, + "output": 0.25 } }, - "unsloth/gemma-3-27b-it": { - "id": "unsloth/gemma-3-27b-it", - "name": "Gemma 3 27B IT", - "family": "unsloth", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "OpenAI: GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": false, - "structured_output": false, - "release_date": "2025-03-10", - "last_updated": "2025-03-10", + "structured_output": true, + "temperature": true, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 96000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.2992, - "output": 0.2992 + "input": 0.04, + "output": 0.15 } }, - "THUDM/GLM-Z1-9B-0414": { - "id": "THUDM/GLM-Z1-9B-0414", - "name": "GLM Z1 9B 0414", - "family": "glm-z", + "sao10K/l31-70b-euryale-v2.2": { + "id": "sao10K/l31-70b-euryale-v2.2", + "name": "L31 70B Euryale V2.2", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": true, + "temperature": true, + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, - "output": 8000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 1.48, + "output": 1.48 } }, - "THUDM/GLM-4-9B-0414": { - "id": "THUDM/GLM-4-9B-0414", - "name": "GLM 4 9B 0414", - "family": "glm", + "sao10K/L3-8B-stheno-v3.2": { + "id": "sao10K/L3-8B-stheno-v3.2", + "name": "L3 8B Stheno V3.2", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": true, + "temperature": true, + "release_date": "2024-11-29", + "last_updated": "2024-11-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, - "output": 8000 + "context": 8192, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.05, + "output": 0.05 } }, - "THUDM/GLM-Z1-Rumination-32B-0414": { - "id": "THUDM/GLM-Z1-Rumination-32B-0414", - "name": "GLM Z1 Rumination 32B 0414", - "family": "glm-z", + "sao10K/l3-8b-lunaris": { + "id": "sao10K/l3-8b-lunaris", + "name": "Sao10k L3 8B Lunaris\t", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "structured_output": true, + "temperature": true, + "release_date": "2024-11-28", + "last_updated": "2024-11-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "input": 32000, - "output": 65536 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.2 - } + "input": 0.05, + "output": 0.05 + } }, - "THUDM/GLM-4-32B-0414": { - "id": "THUDM/GLM-4-32B-0414", - "name": "GLM 4 32B 0414", - "family": "glm", + "sao10K/l3-70b-euryale-v2.1": { + "id": "sao10K/l3-70b-euryale-v2.1", + "name": "L3 70B Euryale V2.1\t", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": true, + "temperature": true, + "release_date": "2024-06-18", + "last_updated": "2024-06-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 1.48, + "output": 1.48 } }, - "THUDM/GLM-Z1-32B-0414": { - "id": "THUDM/GLM-Z1-32B-0414", - "name": "GLM Z1 32B 0414", - "family": "glm-z", + "baichuan/baichuan-m2-32b": { + "id": "baichuan/baichuan-m2-32b", + "name": "baichuan-m2-32b", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "baichuan", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-08-13", + "last_updated": "2025-08-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, - "limit": { - "context": 128000, - "input": 128000, - "output": 65536 - }, - "cost": { - "input": 0.2, - "output": 0.2 - } - }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Gemini 3 Flash (Preview)", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.07, + "output": 0.07 } }, - "google/gemini-flash-1.5": { - "id": "google/gemini-flash-1.5", - "name": "Gemini 1.5 Flash", - "family": "gemini-flash", + "mistralai/mistral-nemo": { + "id": "mistralai/mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-05-14", - "last_updated": "2024-05-14", + "structured_output": true, + "temperature": true, + "release_date": "2024-07-30", + "last_updated": "2024-07-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "input": 2000000, - "output": 8192 + "context": 60288, + "output": 16000 }, "cost": { - "input": 0.0748, - "output": 0.306 + "input": 0.04, + "output": 0.17 } }, - "google/gemini-3-flash-preview-thinking": { - "id": "google/gemini-3-flash-preview-thinking", - "name": "Gemini 3 Flash Thinking", - "family": "gemini-flash", - "attachment": true, + "xiaomimimo/mimo-v2-flash": { + "id": "xiaomimimo/mimo-v2-flash", + "name": "XiaomiMiMo/MiMo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", + "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-12-19", + "last_updated": "2025-12-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048756, - "input": 1048756, - "output": 65536 + "context": 262144, + "output": 32000 }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.1, + "output": 0.3, + "cache_read": 0.3 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, - "reasoning": false, + "xiaomimimo/mimo-v2-pro": { + "id": "xiaomimimo/mimo-v2-pro", + "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, - "release_date": "2026-01-26", - "last_updated": "2026-01-26", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "input": 256000, - "output": 65536 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.9 + "input": 2, + "output": 6, + "cache_read": 0.4, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } } }, - "moonshotai/kimi-k2-instruct": { - "id": "moonshotai/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", + "xiaomimimo/mimo-v2.5-pro": { + "id": "xiaomimimo/mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 8192 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 2 + "input": 0.522, + "output": 1.044, + "cache_read": 0.0043, + "tiers": [ + { + "input": 0.522, + "output": 1.044, + "cache_read": 0.0043, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.522, + "output": 1.044, + "cache_read": 0.0043 + } } }, - "moonshotai/kimi-k2-thinking-original": { - "id": "moonshotai/kimi-k2-thinking-original", - "name": "Kimi K2 Thinking Original", - "family": "kimi-thinking", + "gryphe/mythomax-l2-13b": { + "id": "gryphe/mythomax-l2-13b", + "name": "Mythomax L2 13B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "temperature": true, + "release_date": "2024-04-25", + "last_updated": "2024-04-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 4096, + "output": 3200 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.09, + "output": 0.09 } }, - "moonshotai/kimi-k2-instruct-0711": { - "id": "moonshotai/kimi-k2-instruct-0711", - "name": "Kimi K2 0711", - "family": "kimi", + "zai-org/glm-4.7": { + "id": "zai-org/glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "temperature": true, + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 2 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "moonshotai/Kimi-Dev-72B": { - "id": "moonshotai/Kimi-Dev-72B", - "name": "Kimi Dev 72B", - "family": "kimi", + "zai-org/glm-4.5v": { + "id": "zai-org/glm-4.5v", + "name": "GLM 4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "video", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 131072 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 0.6, + "output": 1.8, + "cache_read": 0.11 } }, - "moonshotai/kimi-k2-thinking-turbo-original": { - "id": "moonshotai/kimi-k2-thinking-turbo-original", - "name": "Kimi K2 Thinking Turbo Original", - "family": "kimi-thinking", + "zai-org/glm-4.5": { + "id": "zai-org/glm-4.5", + "name": "GLM-4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 16384 + "context": 131072, + "output": 98304 }, "cost": { - "input": 1.15, - "output": 8 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "zai-org/autoglm-phone-9b-multilingual": { + "id": "zai-org/autoglm-phone-9b-multilingual", + "name": "AutoGLM-Phone-9B-Multilingual", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": false, - "release_date": "2026-04-16", - "last_updated": "2026-04-21", + "tool_call": false, + "temperature": true, + "release_date": "2025-12-10", + "last_updated": "2025-12-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, + "context": 65536, "output": 65536 }, "cost": { - "input": 0.53, - "output": 2.73 + "input": 0.035, + "output": 0.138 } }, - "moonshotai/kimi-k2.6:thinking": { - "id": "moonshotai/kimi-k2.6:thinking", - "name": "Kimi K2.6 Thinking", - "family": "kimi-thinking", - "attachment": true, + "zai-org/glm-5.1": { + "id": "zai-org/glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, - "release_date": "2026-04-16", - "last_updated": "2026-04-21", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 65536 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.53, - "output": 2.73 + "input": 1.38, + "output": 4.4, + "cache_read": 0.26 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "zai-org/glm-4.6": { + "id": "zai-org/glm-4.6", + "name": "GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 262144 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.55, + "output": 2.2, + "cache_read": 0.11 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "zai-org/glm-5.2": { + "id": "zai-org/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 262144 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "moonshotai/kimi-k2.5:thinking": { - "id": "moonshotai/kimi-k2.5:thinking", - "name": "Kimi K2.5 Thinking", - "family": "kimi-thinking", + "zai-org/glm-4.6v": { + "id": "zai-org/glm-4.6v", + "name": "GLM 4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, - "release_date": "2026-01-26", - "last_updated": "2026-01-26", + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text", "image"], + "input": ["text", "video", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { "input": 0.3, - "output": 1.9 + "output": 0.9, + "cache_read": 0.055 } }, - "Tongyi-Zhiwen/QwenLong-L1-32B": { - "id": "Tongyi-Zhiwen/QwenLong-L1-32B", - "name": "QwenLong L1 32B", - "family": "qwen", + "zai-org/glm-4.5-air": { + "id": "zai-org/glm-4.5-air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-10-13", + "last_updated": "2025-10-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 40960 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.13999999999999999, - "output": 0.6 + "input": 0.13, + "output": 0.85, + "cache_read": 0.025 } }, - "nothingiisreal/L3.1-70B-Celeste-V0.1-BF16": { - "id": "nothingiisreal/L3.1-70B-Celeste-V0.1-BF16", - "name": "Llama 3.1 70B Celeste v0.1", - "family": "llama", + "zai-org/glm-4.7-flash": { + "id": "zai-org/glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.07, + "output": 0.4, + "cache_read": 0.01 } }, - "aion-labs/aion-1.0": { - "id": "aion-labs/aion-1.0", - "name": "Aion 1.0", - "family": "llama", + "zai-org/glm-5": { + "id": "zai-org/glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-02-01", - "last_updated": "2025-02-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "input": 65536, - "output": 8192 + "context": 202800, + "output": 131072 }, "cost": { - "input": 3.995, - "output": 7.99 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "aion-labs/aion-rp-llama-3.1-8b": { - "id": "aion-labs/aion-rp-llama-3.1-8b", - "name": "Llama 3.1 8b (uncensored)", - "family": "llama", - "attachment": false, + "paddlepaddle/paddleocr-vl": { + "id": "paddlepaddle/paddleocr-vl", + "name": "PaddleOCR-VL", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "temperature": true, + "release_date": "2025-10-22", + "last_updated": "2025-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, + "context": 16384, "output": 16384 }, "cost": { - "input": 0.2006, - "output": 0.2006 + "input": 0.02, + "output": 0.02 } }, - "aion-labs/aion-1.0-mini": { - "id": "aion-labs/aion-1.0-mini", - "name": "Aion 1.0 mini (DeepSeek)", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "qwen/qwen3-vl-235b-a22b-thinking": { + "id": "qwen/qwen3-vl-235b-a22b-thinking", + "name": "Qwen3 VL 235B A22B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-02-20", - "last_updated": "2025-02-20", + "temperature": true, + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "input": 131072, - "output": 8192 + "output": 32768 }, "cost": { - "input": 0.7989999999999999, - "output": 1.394 + "input": 0.98, + "output": 3.95 } }, - "Alibaba-NLP/Tongyi-DeepResearch-30B-A3B": { - "id": "Alibaba-NLP/Tongyi-DeepResearch-30B-A3B", - "name": "Tongyi DeepResearch 30B A3B", - "family": "yi", - "attachment": false, + "qwen/qwen3-vl-30b-a3b-thinking": { + "id": "qwen/qwen3-vl-30b-a3b-thinking", + "name": "qwen/qwen3-vl-30b-a3b-thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-11", + "last_updated": "2025-10-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.08, - "output": 0.24000000000000002 + "input": 0.2, + "output": 1 } }, - "MiniMaxAI/MiniMax-M1-80k": { - "id": "MiniMaxAI/MiniMax-M1-80k", - "name": "MiniMax M1 80K", - "family": "minimax", + "qwen/qwen3-235b-a22b-instruct-2507": { + "id": "qwen/qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-06-16", - "last_updated": "2025-06-16", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 131072 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.6052, - "output": 2.4225000000000003 + "input": 0.09, + "output": 0.58 } }, - "anthropic/claude-opus-4.6:thinking:low": { - "id": "anthropic/claude-opus-4.6:thinking:low", - "name": "Claude 4.6 Opus Thinking Low", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "qwen/qwen3-coder-30b-a3b-instruct": { + "id": "qwen/qwen3-coder-30b-a3b-instruct", + "name": "Qwen3 Coder 30b A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "release_date": "2025-10-09", + "last_updated": "2025-10-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 160000, + "output": 32768 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 0.07, + "output": 0.27 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Claude 4.6 Opus", - "family": "claude-opus", - "attachment": true, + "qwen/qwen3-8b-fp8": { + "id": "qwen/qwen3-8b-fp8", + "name": "Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 128000, + "output": 20000 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 0.035, + "output": 0.138 } }, - "anthropic/claude-sonnet-4.6:thinking": { - "id": "anthropic/claude-sonnet-4.6:thinking", - "name": "Claude Sonnet 4.6 Thinking", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "qwen/qwen3-next-80b-a3b-instruct": { + "id": "qwen/qwen3-next-80b-a3b-instruct", + "name": "Qwen3 Next 80B A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "temperature": true, + "release_date": "2025-09-10", + "last_updated": "2025-09-10", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 2.992, - "output": 14.993999999999998 + "input": 0.15, + "output": 1.5 } }, - "anthropic/claude-opus-4.6:thinking:max": { - "id": "anthropic/claude-opus-4.6:thinking:max", - "name": "Claude 4.6 Opus Thinking Max", - "family": "claude-opus", + "qwen/qwen3-vl-8b-instruct": { + "id": "qwen/qwen3-vl-8b-instruct", + "name": "qwen/qwen3-vl-8b-instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "release_date": "2025-10-17", + "last_updated": "2025-10-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 0.08, + "output": 0.5 } }, - "anthropic/claude-opus-4.6:thinking:medium": { - "id": "anthropic/claude-opus-4.6:thinking:medium", - "name": "Claude 4.6 Opus Thinking Medium", - "family": "claude-opus", + "qwen/qwen3-omni-30b-a3b-thinking": { + "id": "qwen/qwen3-omni-30b-a3b-thinking", + "name": "Qwen3 Omni 30B A3B Thinking", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "audio", "video", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 65536, + "output": 16384 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 0.25, + "output": 0.97, + "input_audio": 2.2, + "output_audio": 1.788 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, - "reasoning": false, + "qwen/qwen3.7-max": { + "id": "qwen/qwen3.7-max", + "name": "Qwen3.7-Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "input": 1000000, - "output": 128000 + "output": 65536 }, "cost": { - "input": 2.992, - "output": 14.993999999999998 + "input": 1.25, + "output": 3.75, + "cache_read": 0.25, + "cache_write": 1.5625 } }, - "anthropic/claude-opus-4.6:thinking": { - "id": "anthropic/claude-opus-4.6:thinking", - "name": "Claude 4.6 Opus Thinking", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 4.998, - "output": 25.007 + "input": 2.11, + "output": 8.45 } }, - "abacusai/Dracarys-72B-Instruct": { - "id": "abacusai/Dracarys-72B-Instruct", - "name": "Llama 3.1 70B Dracarys 2", - "family": "llama", + "qwen/qwen2.5-7b-instruct": { + "id": "qwen/qwen2.5-7b-instruct", + "name": "Qwen2.5 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-08-02", - "last_updated": "2025-08-02", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.07, + "output": 0.07 } }, - "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0": { - "id": "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0", - "name": "EVA Llama 3.33 70B", - "family": "llama", + "qwen/qwen3-next-80b-a3b-thinking": { + "id": "qwen/qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-10", + "last_updated": "2025-09-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 2.006, - "output": 2.006 + "input": 0.15, + "output": 1.5 } }, - "EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2": { - "id": "EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2", - "name": "EVA-Qwen2.5-72B-v0.2", + "qwen/qwen3-235b-a22b-thinking-2507": { + "id": "qwen/qwen3-235b-a22b-thinking-2507", + "name": "Qwen3 235B A22b Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.7989999999999999, - "output": 0.7989999999999999 + "input": 0.3, + "output": 3 } }, - "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1": { - "id": "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1", - "name": "EVA-LLaMA-3.33-70B-v0.1", - "family": "llama", + "qwen/qwen-mt-plus": { + "id": "qwen/qwen-mt-plus", + "name": "Qwen MT Plus", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "release_date": "2025-09-03", + "last_updated": "2025-09-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 16384, - "input": 16384, - "output": 16384 + "output": 8192 }, "cost": { - "input": 2.006, - "output": 2.006 + "input": 0.25, + "output": 0.75 } }, - "EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2": { - "id": "EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2", - "name": "EVA-Qwen2.5-32B-v0.2", - "family": "qwen", + "qwen/qwen3-32b-fp8": { + "id": "qwen/qwen3-32b-fp8", + "name": "Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "temperature": true, + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 40960, + "output": 20000 }, "cost": { - "input": 0.7989999999999999, - "output": 0.7989999999999999 + "input": 0.1, + "output": 0.45 } }, - "huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated": { - "id": "huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated", - "name": "DeepSeek R1 Qwen Abliterated", - "family": "qwen", - "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 16384, - "input": 16384, - "output": 8192 - }, - "cost": { - "input": 1.4, - "output": 1.4 - } - }, - "huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated": { - "id": "huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated", - "name": "DeepSeek R1 Llama 70B Abliterated", - "family": "deepseek", + "qwen/qwen3-4b-fp8": { + "id": "qwen/qwen3-4b-fp8", + "name": "Qwen3 4B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "temperature": true, + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 128000, + "output": 20000 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.03, + "output": 0.03 } }, - "huihui-ai/Llama-3.3-70B-Instruct-abliterated": { - "id": "huihui-ai/Llama-3.3-70B-Instruct-abliterated", - "name": "Llama 3.3 70B Instruct abliterated", - "family": "llama", - "attachment": false, + "qwen/qwen2.5-vl-72b-instruct": { + "id": "qwen/qwen2.5-vl-72b-instruct", + "name": "Qwen2.5 VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "temperature": true, + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.8, + "output": 0.8 } }, - "huihui-ai/Qwen2.5-32B-Instruct-abliterated": { - "id": "huihui-ai/Qwen2.5-32B-Instruct-abliterated", - "name": "Qwen 2.5 32B Abliterated", - "family": "qwen", + "qwen/qwen3-30b-a3b-fp8": { + "id": "qwen/qwen3-30b-a3b-fp8", + "name": "Qwen3 30B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": false, - "release_date": "2025-01-06", - "last_updated": "2025-01-06", + "temperature": true, + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 8192 + "context": 40960, + "output": 20000 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.09, + "output": 0.45 } }, - "huihui-ai/Llama-3.1-Nemotron-70B-Instruct-HF-abliterated": { - "id": "huihui-ai/Llama-3.1-Nemotron-70B-Instruct-HF-abliterated", - "name": "Nemotron 3.1 70B abliterated", - "family": "nemotron", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "qwen/qwen3.5-27b": { + "id": "qwen/qwen3.5-27b", + "name": "Qwen3.5-27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.3, + "output": 2.4 } }, - "xiaomi/mimo-v2-flash-thinking-original": { - "id": "xiaomi/mimo-v2-flash-thinking-original", - "name": "MiMo V2 Flash (Thinking) Original", - "family": "mimo", + "qwen/qwen-2.5-72b-instruct": { + "id": "qwen/qwen-2.5-72b-instruct", + "name": "Qwen 2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-10-15", + "last_updated": "2024-10-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 32000, + "output": 8192 }, "cost": { - "input": 0.102, - "output": 0.306 + "input": 0.38, + "output": 0.4 } }, - "xiaomi/mimo-v2-flash-thinking": { - "id": "xiaomi/mimo-v2-flash-thinking", - "name": "MiMo V2 Flash (Thinking)", - "family": "mimo", + "qwen/qwen3-coder-next": { + "id": "qwen/qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.102, - "output": 0.306 + "input": 0.2, + "output": 1.5 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "MiMo V2 Flash", - "family": "mimo", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "qwen/qwen3.5-35b-a3b": { + "id": "qwen/qwen3.5-35b-a3b", + "name": "Qwen3.5-35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.102, - "output": 0.306 + "input": 0.25, + "output": 2 } }, - "xiaomi/mimo-v2-flash-original": { - "id": "xiaomi/mimo-v2-flash-original", - "name": "MiMo V2 Flash Original", - "family": "mimo", + "qwen/qwen3-coder-480b-a35b-instruct": { + "id": "qwen/qwen3-coder-480b-a35b-instruct", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.102, - "output": 0.306 + "input": 0.38, + "output": 1.55 } }, - "tngtech/DeepSeek-TNG-R1T2-Chimera": { - "id": "tngtech/DeepSeek-TNG-R1T2-Chimera", - "name": "DeepSeek TNG R1T2 Chimera", - "family": "tngtech", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5-397B-A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 262144, + "output": 64000 }, "cost": { - "input": 0.31, - "output": 0.31 + "input": 0.6, + "output": 3.6 } }, - "tngtech/tng-r1t-chimera": { - "id": "tngtech/tng-r1t-chimera", - "name": "TNG R1T Chimera", - "family": "tngtech", - "attachment": false, + "qwen/qwen3-vl-30b-a3b-instruct": { + "id": "qwen/qwen3-vl-30b-a3b-instruct", + "name": "qwen/qwen3-vl-30b-a3b-instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-11-26", - "last_updated": "2025-11-26", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-11", + "last_updated": "2025-10-11", "modalities": { - "input": ["text"], + "input": ["text", "video", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.2, + "output": 0.7 } }, - "inflatebot/MN-12B-Mag-Mell-R1": { - "id": "inflatebot/MN-12B-Mag-Mell-R1", - "name": "Mag Mell R1", - "family": "mistral-nemo", - "attachment": false, + "qwen/qwen3-omni-30b-a3b-instruct": { + "id": "qwen/qwen3-omni-30b-a3b-instruct", + "name": "Qwen3 Omni 30B A3B Instruct", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "video", "audio", "image"], + "output": ["text", "audio"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.49299999999999994, - "output": 0.49299999999999994 + "input": 0.25, + "output": 0.97, + "input_audio": 2.2, + "output_audio": 1.788 } }, - "failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5": { - "id": "failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5", - "name": "Llama 3 70B abliterated", - "family": "llama", - "attachment": false, + "qwen/qwen3-vl-235b-a22b-instruct": { + "id": "qwen/qwen3-vl-235b-a22b-instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "input": 8192, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.3, + "output": 1.5 } - } - } - }, - "abacus": { - "id": "abacus", - "env": ["ABACUS_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://routellm.abacus.ai/v1", - "name": "Abacus", - "doc": "https://abacus.ai/help/api", - "models": { - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", - "family": "gpt", + }, + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen3.5-122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.4, + "output": 3.2 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "qwen/qwen3-235b-a22b-fp8": { + "id": "qwen/qwen3-235b-a22b-fp8", + "name": "Qwen3 235B A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 40960, + "output": 20000 }, "cost": { - "input": 5, - "output": 25 + "input": 0.2, + "output": 0.8 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "deepseek/deepseek-r1-0528": { + "id": "deepseek/deepseek-r1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 163840, "output": 32768 }, "cost": { - "input": 0.6, - "output": 3 + "input": 0.7, + "output": 2.5, + "cache_read": 0.35 } }, - "gemini-3.1-flash-lite-preview": { - "id": "gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash", - "attachment": true, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2026-03-01", - "last_updated": "2026-03-01", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 1048576, - "output": 65536 + "output": 393216 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "cache_write": 1 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-v3.1-terminus": { + "id": "deepseek/deepseek-v3.1-terminus", + "name": "Deepseek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 3, - "output": 15 + "input": 0.27, + "output": 1, + "cache_read": 0.135 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "deepseek/deepseek-v3-0324": { + "id": "deepseek/deepseek-v3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "knowledge": "2024-07", + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 163840, + "output": 163840 }, "cost": { - "input": 2, - "output": 12 + "input": 0.27, + "output": 1.12, + "cache_read": 0.135 } }, - "gpt-5.3-chat-latest": { - "id": "gpt-5.3-chat-latest", - "name": "GPT-5.3 Chat Latest", - "family": "gpt", + "deepseek/deepseek-ocr": { + "id": "deepseek/deepseek-ocr", + "name": "DeepSeek-OCR", + "description": "OCR model for extracting structured text from documents and screenshots", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2026-03-01", - "last_updated": "2026-03-01", + "release_date": "2025-10-24", + "last_updated": "2025-10-24", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.03, + "output": 0.03 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", - "attachment": true, + "deepseek/deepseek-r1-distill-llama-70b": { + "id": "deepseek/deepseek-r1-distill-llama-70b", + "name": "DeepSeek R1 Distill LLama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.8, + "output": 0.8 } }, - "llama-3.3-70b-versatile": { - "id": "llama-3.3-70b-versatile", - "name": "Llama 3.3 70B Versatile", - "family": "llama", + "deepseek/deepseek-r1-turbo": { + "id": "deepseek/deepseek-r1-turbo", + "name": "DeepSeek R1 (Turbo)\t", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 64000, + "output": 16000 }, "cost": { - "input": 0.59, - "output": 0.79 + "input": 0.7, + "output": 2.5 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "deepseek/deepseek-prover-v2-671b": { + "id": "deepseek/deepseek-prover-v2-671b", + "name": "Deepseek Prover V2 671B", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 160000, + "output": 160000 }, "cost": { - "input": 0.25, - "output": 2 + "input": 0.7, + "output": 2.5 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.05, - "output": 0.4 - } - }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048576, + "output": 393216 }, "cost": { - "input": 1.75, - "output": 14 + "input": 1.6, + "output": 3.2, + "cache_read": 0.135 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-v3.2-exp": { + "id": "deepseek/deepseek-v3.2-exp", + "name": "Deepseek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", "release_date": "2025-09-29", "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 163840, + "output": 65536 }, "cost": { - "input": 3, - "output": 15 + "input": 0.27, + "output": 0.41 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + "deepseek/deepseek-ocr-2": { + "id": "deepseek/deepseek-ocr-2", + "name": "deepseek/deepseek-ocr-2", + "description": "OCR model for extracting structured text from documents and screenshots", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "reasoning": false, + "tool_call": false, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 8192, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.03, + "output": 0.03 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, + "deepseek/deepseek-r1-distill-qwen-14b": { + "id": "deepseek/deepseek-r1-distill-qwen-14b", + "name": "DeepSeek R1 Distill Qwen 14B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "deepseek-thinking", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2025-11-17", - "last_updated": "2025-11-17", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, + "context": 32768, "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.15, + "output": 0.15 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, + "deepseek/deepseek-v3.1": { + "id": "deepseek/deepseek-v3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "structured_output": true, + "temperature": true, + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.27, + "output": 1, + "cache_read": 0.135 } }, - "o3-pro": { - "id": "o3-pro", - "name": "o3-pro", - "family": "o-pro", - "attachment": true, + "deepseek/deepseek-r1-0528-qwen3-8b": { + "id": "deepseek/deepseek-r1-0528-qwen3-8b", + "name": "DeepSeek R1 0528 Qwen3 8B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-06-10", - "last_updated": "2025-06-10", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-05-29", + "last_updated": "2025-05-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "output": 32000 }, "cost": { - "input": 20, - "output": 40 + "input": 0.06, + "output": 0.09 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "GPT-4o Mini", - "family": "gpt", - "attachment": true, + "deepseek/deepseek-r1-distill-qwen-32b": { + "id": "deepseek/deepseek-r1-distill-qwen-32b", + "name": "DeepSeek R1 Distill Qwen 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "deepseek-thinking", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 64000, + "output": 32000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.3, + "output": 0.3 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3 Max", - "family": "qwen", + "deepseek/deepseek-v3-turbo": { + "id": "deepseek/deepseek-v3-turbo", + "name": "DeepSeek V3 (Turbo)\t", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 64000, + "output": 16000 }, "cost": { - "input": 1.2, - "output": 6 + "input": 0.4, + "output": 1.3 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", - "attachment": true, + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "Deepseek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 163840, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 0.269, + "output": 0.4, + "cache_read": 0.1345 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt", - "attachment": true, + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 204800, + "output": 131100 }, "cost": { "input": 0.3, - "output": 2.5 + "output": 1.2, + "cache_read": 0.03 } }, - "gpt-5.2-chat-latest": { - "id": "gpt-5.2-chat-latest", - "name": "GPT-5.2 Chat Latest", - "family": "gpt", - "attachment": true, - "reasoning": true, + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "Minimax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2026-01-01", - "last_updated": "2026-01-01", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 1.75, - "output": 14 - } - }, - "gpt-5.3-codex-xhigh": { - "id": "gpt-5.3-codex-xhigh", - "name": "GPT-5.3 Codex XHigh", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "cost": { - "input": 1.75, - "output": 14 - } - }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", - "attachment": true, - "reasoning": false, - "tool_call": true, "temperature": true, - "release_date": "2025-09-01", - "last_updated": "2025-09-01", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o-mini", + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax-m2.7", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "grok-4-0709": { - "id": "grok-4-0709", - "name": "Grok 4", - "family": "grok", - "attachment": true, + "minimax/minimax-m2.5-highspeed": { + "id": "minimax/minimax-m2.5-highspeed", + "name": "MiniMax M2.5 Highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax-m2.5", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 204800, + "output": 131100 }, "cost": { - "input": 3, - "output": 15 + "input": 0.6, + "output": 2.4, + "cache_read": 0.03 } }, - "route-llm": { - "id": "route-llm", - "name": "Route LLM", - "family": "gpt", - "attachment": true, + "kwaipilot/kat-coder-pro": { + "id": "kwaipilot/kat-coder-pro", + "name": "Kat Coder Pro", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-01-01", - "last_updated": "2024-01-01", + "release_date": "2026-01-05", + "last_updated": "2026-01-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 256000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "qwen-2.5-coder-32b": { - "id": "qwen-2.5-coder-32b", - "name": "Qwen 2.5 Coder 32B", - "family": "qwen", + "nousresearch/hermes-2-pro-llama-3-8b": { + "id": "nousresearch/hermes-2-pro-llama-3-8b", + "name": "Hermes 2 Pro Llama 3 8B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2024-11-11", - "last_updated": "2024-11-11", + "release_date": "2024-06-27", + "last_updated": "2024-06-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, + "context": 8192, "output": 8192 }, "cost": { - "input": 0.79, - "output": 0.79 + "input": 0.14, + "output": 0.14 } - }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "GPT-5 Codex", - "family": "gpt", - "attachment": false, + } + } + }, + "xai": { + "id": "xai", + "env": ["XAI_API_KEY"], + "npm": "@ai-sdk/xai", + "name": "xAI", + "doc": "https://docs.x.ai/docs/models", + "models": { + "grok-4.20-multi-agent-0309": { + "id": "grok-4.20-multi-agent-0309", + "name": "Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "temperature": true, + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1000000, + "output": 30000 }, "cost": { "input": 1.25, - "output": 10 + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "grok-4.20-0309-non-reasoning": { + "id": "grok-4.20-0309-non-reasoning", + "name": "Grok 4.20 (Non-Reasoning)", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 30000 }, "cost": { - "input": 15, - "output": 75 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + "grok-4.3": { + "id": "grok-4.3", + "name": "Grok 4.3", + "description": "xAI's Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1000000, + "output": 30000 }, "cost": { - "input": 2.5, - "output": 15 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "gpt-5.1-chat-latest": { - "id": "gpt-5.1-chat-latest", - "name": "GPT-5.1 Chat Latest", - "family": "gpt", + "grok-imagine-image-quality": { + "id": "grok-imagine-image-quality", + "name": "Grok Imagine Image Quality", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-04-03", + "last_updated": "2026-04-03", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["image", "pdf"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.25, - "output": 10 + "context": 8000, + "output": 0 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "grok-imagine-video": { + "id": "grok-imagine-video", + "name": "Grok Imagine Video", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-01-28", + "last_updated": "2026-01-28", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image", "video", "pdf"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 1, - "output": 5 + "context": 1024, + "output": 0 } }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "grok-4.5": { + "id": "grok-4.5", + "name": "Grok 4.5", + "description": "xAI's latest Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-05-14", - "last_updated": "2025-05-14", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15 - } - }, - "kimi-k2-turbo-preview": { - "id": "kimi-k2-turbo-preview", - "name": "Kimi K2 Turbo Preview", - "family": "kimi", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 256000, - "output": 8192 + "context": 500000, + "output": 500000 }, "cost": { - "input": 0.15, - "output": 8 + "input": 2, + "output": 6, + "cache_read": 0.5, + "tiers": [ + { + "input": 4, + "output": 12, + "cache_read": 1, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 12, + "cache_read": 1 + } } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "grok-4.20-0309-reasoning": { + "id": "grok-4.20-0309-reasoning", + "name": "Grok 4.20 (Reasoning)", + "description": "Reasoning Grok for document-heavy analysis and long-horizon tool use", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 1000000, + "output": 30000 }, "cost": { - "input": 5, - "output": 25 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "GPT-4.1 Nano", - "family": "gpt", + "grok-imagine-image": { + "id": "grok-imagine-image", + "name": "Grok Imagine Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "temperature": false, + "release_date": "2026-01-28", + "last_updated": "2026-01-28", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["image", "pdf"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 - }, - "cost": { - "input": 0.1, - "output": 0.4 + "context": 8000, + "output": 0 } }, - "claude-3-7-sonnet-20250219": { - "id": "claude-3-7-sonnet-20250219", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", + "grok-build-0.1": { + "id": "grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 3, - "output": 15 + "input": 1, + "output": 2, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 4, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 4, + "cache_read": 0.4 + } } - }, - "o3": { - "id": "o3", - "name": "o3", - "family": "o", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + } + } + }, + "privatemode-ai": { + "id": "privatemode-ai", + "env": ["PRIVATEMODE_API_KEY", "PRIVATEMODE_ENDPOINT"], + "npm": "@ai-sdk/openai-compatible", + "api": "http://localhost:8080/v1", + "name": "Privatemode AI", + "doc": "https://docs.privatemode.ai/api/overview", + "models": { + "qwen3-embedding-4b": { + "id": "qwen3-embedding-4b", + "name": "Qwen3-Embedding 4B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-06-06", + "last_updated": "2025-06-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 32000, + "output": 2560 }, "cost": { - "input": 2, - "output": 8 + "input": 0, + "output": 0 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", + "gemma-3-27b": { + "id": "gemma-3-27b", + "name": "Gemma 3 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0, + "output": 0 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-05-14", - "last_updated": "2025-05-14", + "knowledge": "2025-08", + "release_date": "2025-08-04", + "last_updated": "2025-08-14", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 15, - "output": 75 + "input": 0, + "output": 0 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "whisper-large-v3": { + "id": "whisper-large-v3", + "name": "Whisper large-v3", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2023-09", + "release_date": "2023-09-01", + "last_updated": "2023-09-01", "modalities": { - "input": ["text", "image"], + "input": ["audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 0, + "output": 4096 }, "cost": { - "input": 2, - "output": 8 + "input": 0, + "output": 0 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 Mini", - "family": "gpt", - "attachment": true, + "qwen3-coder-30b-a3b": { + "id": "qwen3-coder-30b-a3b", + "name": "Qwen3-Coder 30B-A3B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, + "context": 128000, "output": 32768 }, "cost": { - "input": 0.4, - "output": 1.6 + "input": 0, + "output": 0 } - }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt", - "attachment": true, - "reasoning": true, + } + } + }, + "drun": { + "id": "drun", + "env": ["DRUN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://chat.d.run/v1", + "name": "D.Run (China)", + "doc": "https://www.d.run", + "models": { + "public/deepseek-v3": { + "id": "public/deepseek-v3", + "name": "DeepSeek V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-12-26", + "last_updated": "2024-12-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.28, + "output": 1.1 } }, - "gpt-4o-2024-11-20": { - "id": "gpt-4o-2024-11-20", - "name": "GPT-4o (2024-11-20)", - "family": "gpt", - "attachment": true, - "reasoning": false, + "public/deepseek-r1": { + "id": "public/deepseek-r1", + "name": "DeepSeek R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-11-20", - "last_updated": "2024-11-20", + "knowledge": "2024-12", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 32000 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.55, + "output": 2.2 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, - "reasoning": false, + "public/minimax-m25": { + "id": "public/minimax-m25", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, "temperature": true, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "release_date": "2025-03-01", + "last_updated": "2025-03-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0.29, + "output": 1.16 } - }, - "deepseek/deepseek-v3.1": { - "id": "deepseek/deepseek-v3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + } + } + }, + "alibaba-token-plan-cn": { + "id": "alibaba-token-plan-cn", + "env": ["ALIBABA_TOKEN_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1", + "name": "Alibaba Token Plan (China)", + "doc": "https://www.alibabacloud.com/help/zh/model-studio/token-plan-overview", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.55, - "output": 1.66 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "Qwen/QwQ-32B": { - "id": "Qwen/QwQ-32B", - "name": "QwQ 32B", + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-11-28", - "last_updated": "2024-11-28", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct", + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.13, - "output": 0.6 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "Qwen/Qwen3-32B": { - "id": "Qwen/Qwen3-32B", - "name": "Qwen3 32B", - "family": "qwen", - "attachment": false, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.09, - "output": 0.29 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "Qwen/qwen3-coder-480b-a35b-instruct": { - "id": "Qwen/qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 202752, + "output": 128000 }, "cost": { - "input": 0.29, - "output": 1.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "Qwen/Qwen2.5-72B-Instruct": { - "id": "Qwen/Qwen2.5-72B-Instruct", - "name": "Qwen 2.5 72B Instruct", - "family": "qwen", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.11, - "output": 0.38 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "zai-org/glm-4.7": { - "id": "zai-org/glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "wan2.7-image-pro": { + "id": "wan2.7-image-pro", + "name": "Wan2.7 Image Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 8192, + "output": 0 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0, + "output": 0 } }, - "zai-org/glm-5": { - "id": "zai-org/glm-5", - "name": "GLM-5", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 1000000, "output": 131072 }, "cost": { - "input": 1, - "output": 3.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/glm-4.5": { - "id": "zai-org/glm-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/glm-4.6": { - "id": "zai-org/glm-4.6", - "name": "GLM-4.6", - "family": "glm", - "attachment": false, - "reasoning": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo": { - "id": "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", - "name": "Llama 3.1 405B Instruct Turbo", - "family": "llama", + "qwen-image-2.0": { + "id": "qwen-image-2.0", + "name": "Qwen Image 2.0", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": false, "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 196608, + "input": 196601, + "output": 24576 }, "cost": { - "input": 3.5, - "output": 3.5 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { - "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", - "name": "Llama 4 Maverick 17B 128E Instruct FP8", - "family": "llama", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 32768 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.59 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "meta-llama/Meta-Llama-3.1-8B-Instruct": { - "id": "meta-llama/Meta-Llama-3.1-8B-Instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "qwen-image-2.0-pro": { + "id": "qwen-image-2.0-pro", + "name": "Qwen Image 2.0 Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 8192, + "output": 0 }, "cost": { - "input": 0.02, - "output": 0.05 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-R1": { - "id": "deepseek-ai/DeepSeek-R1", - "name": "DeepSeek R1", - "family": "deepseek-thinking", + "wan2.7-image": { + "id": "wan2.7-image", + "name": "Wan2.7 Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 8192, + "output": 0 }, "cost": { - "input": 3, - "output": 7 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-06-15", - "last_updated": "2025-06-15", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.27, - "output": 0.4 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "DeepSeek V3.1 Terminus", + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "knowledge": "2025-01", + "release_date": "2025-12-03", + "last_updated": "2025-12-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.27, - "output": 1 + "input": 0, + "output": 0 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT-OSS 120B", - "family": "gpt-oss", - "attachment": true, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.08, - "output": 0.44 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } } } }, - "perplexity-agent": { - "id": "perplexity-agent", - "env": ["PERPLEXITY_API_KEY"], - "npm": "@ai-sdk/openai", - "api": "https://api.perplexity.ai/v1", - "name": "Perplexity Agent", - "doc": "https://docs.perplexity.ai/docs/agent-api/models", + "moonshotai": { + "id": "moonshotai", + "env": ["MOONSHOT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.moonshot.ai/v1", + "name": "Moonshot AI", + "doc": "https://platform.moonshot.ai/docs/api/chat", "models": { - "perplexity/sonar": { - "id": "perplexity/sonar", - "name": "Sonar", - "family": "sonar", + "kimi-k2-0905-preview": { + "id": "kimi-k2-0905-preview", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.25, + "input": 0.6, "output": 2.5, - "cache_read": 0.0625 - } - }, - "xai/grok-4-1-fast-non-reasoning": { - "id": "xai/grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 2000000, - "output": 30000 - }, - "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "cache_read": 0.15 } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "Nemotron 3 Super 120B", - "family": "nemotron", + "kimi-k2-thinking-turbo": { + "id": "kimi-k2-thinking-turbo", + "name": "Kimi K2 Thinking Turbo", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2026-02", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.25, - "output": 2.5 + "input": 1.15, + "output": 8, + "cache_read": 0.15 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", - "attachment": true, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, - "reasoning": true, + "kimi-k2-0711-preview": { + "id": "kimi-k2-0711-preview", + "name": "Kimi K2 0711", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-07-14", + "last_updated": "2025-07-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, - "reasoning": true, + "kimi-k2-turbo-preview": { + "id": "kimi-k2-turbo-preview", + "name": "Kimi K2 Turbo", + "description": "Fast Kimi model for responsive chat, coding help, and agent loops", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.25, + "input": 2.4, "output": 10, - "cache_read": 0.125 + "cache_read": 0.6 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "google/gemini-3.1-pro-preview": { - "id": "google/gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "kimi-k2.7-code-highspeed": { + "id": "kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code HighSpeed", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "tiers": [ - { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 - } + "input": 1.9, + "output": 8, + "cache_read": 0.38 } - }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + } + } + }, + "fireworks-ai": { + "id": "fireworks-ai", + "env": ["FIREWORKS_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.fireworks.ai/inference/v1/", + "name": "Fireworks AI", + "doc": "https://fireworks.ai/docs/", + "models": { + "accounts/fireworks/routers/kimi-k2p6-turbo": { + "id": "accounts/fireworks/routers/kimi-k2p6-turbo", + "name": "Kimi K2.6 Turbo", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - } + "input": 2, + "output": 8, + "cache_read": 0.3 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, + "accounts/fireworks/routers/glm-5p2-fast": { + "id": "accounts/fireworks/routers/glm-5p2-fast", + "name": "GLM 5.2 Fast", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2026-06-26", + "last_updated": "2026-06-26", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 1048575, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03 + "input": 2.1, + "output": 6.6, + "cache_read": 0.21 } }, - "anthropic/claude-haiku-4-5": { - "id": "anthropic/claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "accounts/fireworks/routers/kimi-k2p7-code-fast": { + "id": "accounts/fireworks/routers/kimi-k2p7-code-fast", + "name": "Kimi K2.7 Code Fast", + "description": "Kimi coding model for software agents, refactors, and repository reasoning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-06-12", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1 + "input": 1.9, + "output": 8, + "cache_read": 0.38 } }, - "anthropic/claude-sonnet-4-6": { - "id": "anthropic/claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "accounts/fireworks/routers/glm-5p1-fast": { + "id": "accounts/fireworks/routers/glm-5p1-fast", + "name": "GLM 5.1 Fast", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 202800, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3 + "input": 2.8, + "output": 8.8, + "cache_read": 0.52 } }, - "anthropic/claude-opus-4-7": { - "id": "anthropic/claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "accounts/fireworks/routers/kimi-k2p6-fast": { + "id": "accounts/fireworks/routers/kimi-k2p6-fast", + "name": "Kimi K2.6 Fast", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-06-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5 + "input": 2, + "output": 8, + "cache_read": 0.3 } }, - "anthropic/claude-opus-4-5": { - "id": "anthropic/claude-opus-4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "accounts/fireworks/models/deepseek-v4-flash": { + "id": "accounts/fireworks/models/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "anthropic/claude-opus-4-6": { - "id": "anthropic/claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "accounts/fireworks/models/deepseek-v4-pro": { + "id": "accounts/fireworks/models/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5 + "input": 1.74, + "output": 3.48, + "cache_read": 0.145 } }, - "anthropic/claude-sonnet-4-5": { - "id": "anthropic/claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "accounts/fireworks/models/minimax-m2p7": { + "id": "accounts/fireworks/models/minimax-m2p7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-04-12", + "last_updated": "2026-04-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 196608, + "output": 196608 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } - } - } - }, - "siliconflow-cn": { - "id": "siliconflow-cn", - "env": ["SILICONFLOW_CN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.siliconflow.cn/v1", - "name": "SiliconFlow (China)", - "doc": "https://cloud.siliconflow.com/models", - "models": { - "Kwaipilot/KAT-Dev": { - "id": "Kwaipilot/KAT-Dev", - "name": "Kwaipilot/KAT-Dev", - "family": "kat-coder", - "attachment": false, - "reasoning": false, + }, + "accounts/fireworks/models/minimax-m3": { + "id": "accounts/fireworks/models/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-09-27", - "last_updated": "2026-01-16", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 512000, + "output": 512000 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "Qwen/Qwen3.5-397B-A17B": { - "id": "Qwen/Qwen3.5-397B-A17B", - "name": "Qwen/Qwen3.5-397B-A17B", - "family": "qwen", - "attachment": false, + "accounts/fireworks/models/kimi-k2p6": { + "id": "accounts/fireworks/models/kimi-k2p6", + "name": "Kimi K2.6", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.29, - "output": 1.74 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "Qwen/Qwen3.5-35B-A3B": { - "id": "Qwen/Qwen3.5-35B-A3B", - "name": "Qwen/Qwen3.5-35B-A3B", + "accounts/fireworks/models/qwen3p7-plus": { + "id": "accounts/fireworks/models/qwen3p7-plus", + "name": "Qwen 3.7 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-25", - "last_updated": "2026-02-25", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, "output": 65536 }, "cost": { - "input": 0.23, - "output": 1.86 + "input": 0.4, + "output": 1.6, + "cache_read": 0.08 } }, - "Qwen/Qwen3.5-122B-A10B": { - "id": "Qwen/Qwen3.5-122B-A10B", - "name": "Qwen/Qwen3.5-122B-A10B", - "family": "qwen", + "accounts/fireworks/models/glm-5p1": { + "id": "accounts/fireworks/models/glm-5p1", + "name": "GLM 5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 202800, + "output": 131072 }, "cost": { - "input": 0.29, - "output": 2.32 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "Qwen/Qwen3.5-9B": { - "id": "Qwen/Qwen3.5-9B", - "name": "Qwen/Qwen3.5-9B", - "family": "qwen", + "accounts/fireworks/models/glm-5p2": { + "id": "accounts/fireworks/models/glm-5p2", + "name": "GLM 5.2", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2026-06-16", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 1048575, + "output": 131072 }, "cost": { - "input": 0.22, - "output": 1.74 + "input": 1.4, + "output": 4.4, + "cache_read": 0.14 } }, - "Qwen/Qwen3.5-27B": { - "id": "Qwen/Qwen3.5-27B", - "name": "Qwen/Qwen3.5-27B", - "family": "qwen", + "accounts/fireworks/models/gpt-oss-120b": { + "id": "accounts/fireworks/models/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-25", - "last_updated": "2026-02-25", + "release_date": "2025-08-05", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.26, - "output": 2.09 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "Qwen/Qwen3.5-4B": { - "id": "Qwen/Qwen3.5-4B", - "name": "Qwen/Qwen3.5-4B", - "family": "qwen", + "accounts/fireworks/models/gpt-oss-20b": { + "id": "accounts/fireworks/models/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.07, + "output": 0.3, + "cache_read": 0.035 } }, - "Qwen/Qwen3.6-35B-A3B": { - "id": "Qwen/Qwen3.6-35B-A3B", - "name": "Qwen/Qwen3.6-35B-A3B", - "family": "qwen", - "attachment": false, + "accounts/fireworks/models/kimi-k2p7-code": { + "id": "accounts/fireworks/models/kimi-k2p7-code", + "name": "Kimi K2.7 Code", + "description": "Kimi coding model for software agents, refactors, and repository reasoning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "release_date": "2026-06-12", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.23, - "output": 1.86 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } - }, - "Qwen/Qwen2.5-72B-Instruct": { - "id": "Qwen/Qwen2.5-72B-Instruct", - "name": "Qwen/Qwen2.5-72B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + } + } + }, + "vultr": { + "id": "vultr", + "env": ["VULTR_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.vultrinference.com/v1", + "name": "Vultr", + "doc": "https://api.vultrinference.com/", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.59, - "output": 0.59 + "input": 0.3, + "output": 1.2 } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, - "reasoning": false, + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-07-31", - "last_updated": "2025-11-25", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.3, + "output": 2 } }, - "Qwen/Qwen3-VL-8B-Instruct": { - "id": "Qwen/Qwen3-VL-8B-Instruct", - "name": "Qwen/Qwen3-VL-8B-Instruct", + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.18, - "output": 0.68 + "input": 0.3, + "output": 2 } }, - "Qwen/Qwen3-VL-32B-Instruct": { - "id": "Qwen/Qwen3-VL-32B-Instruct", - "name": "Qwen/Qwen3-VL-32B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "XiaomiMiMo/MiMo-V2.5-Pro": { + "id": "XiaomiMiMo/MiMo-V2.5-Pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-21", - "last_updated": "2025-11-25", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.55, + "output": 1.65 } }, - "Qwen/Qwen3-VL-30B-A3B-Thinking": { - "id": "Qwen/Qwen3-VL-30B-A3B-Thinking", - "name": "Qwen/Qwen3-VL-30B-A3B-Thinking", - "family": "qwen", - "attachment": true, + "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16": { + "id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16", + "name": "NVIDIA Nemotron 3 Nano Omni", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2025-11-25", + "knowledge": "2025-05", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.29, - "output": 1 + "input": 0.13, + "output": 0.38 } }, - "Qwen/Qwen2.5-14B-Instruct": { - "id": "Qwen/Qwen2.5-14B-Instruct", - "name": "Qwen/Qwen2.5-14B-Instruct", - "family": "qwen", + "nvidia/DeepSeek-V3.2-NVFP4": { + "id": "nvidia/DeepSeek-V3.2-NVFP4", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.55, + "output": 1.65 } }, - "Qwen/Qwen3-VL-235B-A22B-Instruct": { - "id": "Qwen/Qwen3-VL-235B-A22B-Instruct", - "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "nvidia/Nemotron-Cascade-2-30B-A3B": { + "id": "nvidia/Nemotron-Cascade-2-30B-A3B", + "name": "NVIDIA Nemotron Cascade 2", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.15, + "output": 0.6 } }, - "Qwen/Qwen3-Next-80B-A3B-Thinking": { - "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "name": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "family": "qwen", + "zai-org/GLM-5.2-FP8": { + "id": "zai-org/GLM-5.2-FP8", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2025-11-25", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 393216, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.85, + "output": 3.1 } }, - "Qwen/Qwen2.5-VL-32B-Instruct": { - "id": "Qwen/Qwen2.5-VL-32B-Instruct", - "name": "Qwen/Qwen2.5-VL-32B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-03-24", - "last_updated": "2025-11-25", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.27, - "output": 0.27 + "input": 0.3, + "output": 1 } }, - "Qwen/Qwen3-Omni-30B-A3B-Thinking": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Thinking", - "name": "Qwen/Qwen3-Omni-30B-A3B-Thinking", - "family": "qwen", - "attachment": true, + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 66000, - "output": 66000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.3, + "output": 1.2 } - }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "family": "qwen", - "attachment": false, + } + } + }, + "302ai": { + "id": "302ai", + "env": ["302AI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.302.ai/v1", + "name": "302.AI", + "doc": "https://doc.302.ai", + "models": { + "gpt-5.4-mini-2026-03-17": { + "id": "gpt-5.4-mini-2026-03-17", + "name": "gpt-5.4-mini-2026-03-17", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.13, - "output": 0.6 + "input": 0.75, + "output": 4.5 } }, - "Qwen/Qwen2.5-32B-Instruct": { - "id": "Qwen/Qwen2.5-32B-Instruct", - "name": "Qwen/Qwen2.5-32B-Instruct", - "family": "qwen", - "attachment": false, + "chatgpt-4o-latest": { + "id": "chatgpt-4o-latest", + "name": "chatgpt-4o-latest", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2024-09-19", - "last_updated": "2025-11-25", + "knowledge": "2023-09", + "release_date": "2024-08-08", + "last_updated": "2024-08-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 5, + "output": 15 } }, - "Qwen/Qwen2.5-72B-Instruct-128K": { - "id": "Qwen/Qwen2.5-72B-Instruct-128K", - "name": "Qwen/Qwen2.5-72B-Instruct-128K", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4-nano-2026-03-17": { + "id": "gpt-5.4-nano-2026-03-17", + "name": "gpt-5.4-nano-2026-03-17", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 4000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.59, - "output": 0.59 + "input": 0.2, + "output": 1.25 } }, - "Qwen/Qwen3-14B": { - "id": "Qwen/Qwen3-14B", - "name": "Qwen/Qwen3-14B", - "family": "qwen", + "kimi-k2-0905-preview": { + "id": "kimi-k2-0905-preview", + "name": "kimi-k2-0905-preview", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "knowledge": "2025-06", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.632, + "output": 2.53 } }, - "Qwen/Qwen3-Omni-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Instruct", - "name": "Qwen/Qwen3-Omni-30B-A3B-Instruct", - "family": "qwen", + "grok-4.20-beta-0309-non-reasoning": { + "id": "grok-4.20-beta-0309-non-reasoning", + "name": "grok-4.20-beta-0309-non-reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 2, + "output": 6 } }, - "Qwen/Qwen3-Coder-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", - "name": "Qwen/Qwen3-Coder-30B-A3B-Instruct", - "family": "qwen", - "attachment": false, + "gemini-2.5-flash-nothink": { + "id": "gemini-2.5-flash-nothink", + "name": "gemini-2.5-flash-nothink", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-08-01", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2025-06-24", + "last_updated": "2025-06-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.3, + "output": 2.5 } }, - "Qwen/Qwen3-32B": { - "id": "Qwen/Qwen3-32B", - "name": "Qwen/Qwen3-32B", + "qwen-plus": { + "id": "qwen-plus", + "name": "Qwen-Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "knowledge": "2024-10", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.12, + "output": 1.2 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "family": "qwen", + "glm-4.7": { + "id": "glm-4.7", + "name": "glm-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.09, - "output": 0.6 + "input": 0.286, + "output": 1.142 } }, - "Qwen/Qwen3-30B-A3B-Instruct-2507": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "family": "qwen", + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "qwen3-235b-a22b-instruct-2507", + "description": "Tool-capable chat model for instruction following and agentic application workflows", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, + "knowledge": "2025-04", "release_date": "2025-07-30", - "last_updated": "2025-11-25", + "last_updated": "2025-07-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 65536 }, "cost": { - "input": 0.09, - "output": 0.3 + "input": 0.29, + "output": 1.143 } }, - "Qwen/Qwen3-8B": { - "id": "Qwen/Qwen3-8B", - "name": "Qwen/Qwen3-8B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "glm-4.5v": { + "id": "glm-4.5v", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-08-12", + "last_updated": "2025-08-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 64000, + "output": 16384 }, "cost": { - "input": 0.06, - "output": 0.06 + "input": 0.29, + "output": 0.86 } }, - "Qwen/Qwen3-Next-80B-A3B-Instruct": { - "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "name": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "claude-opus-4-5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-18", + "knowledge": "2025-03-31", + "release_date": "2025-11-25", "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 5, + "output": 25 } }, - "Qwen/Qwen3-VL-8B-Thinking": { - "id": "Qwen/Qwen3-VL-8B-Thinking", - "name": "Qwen/Qwen3-VL-8B-Thinking", - "family": "qwen", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "gemini-2.5-pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.18, - "output": 2 + "input": 1.25, + "output": 10 } }, - "Qwen/Qwen3-Omni-30B-A3B-Captioner": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Captioner", - "name": "Qwen/Qwen3-Omni-30B-A3B-Captioner", - "family": "qwen", + "gpt-5": { + "id": "gpt-5", + "name": "gpt-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 1.25, + "output": 10 } }, - "Qwen/QwQ-32B": { - "id": "Qwen/QwQ-32B", - "name": "Qwen/QwQ-32B", - "family": "qwen", - "attachment": false, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "claude-haiku-4-5-20251001", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-03-06", - "last_updated": "2025-11-25", + "knowledge": "2025-02-28", + "release_date": "2025-10-16", + "last_updated": "2025-10-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 0.58 + "input": 1, + "output": 5 } }, - "Qwen/Qwen3-VL-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", - "name": "Qwen/Qwen3-VL-30B-A3B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "kimi-k2-thinking-turbo": { + "id": "kimi-k2-thinking-turbo", + "name": "kimi-k2-thinking-turbo", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-05", - "last_updated": "2025-11-25", + "knowledge": "2025-06", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.29, - "output": 1 + "input": 1.265, + "output": 9.119 } }, - "Qwen/Qwen2.5-Coder-32B-Instruct": { - "id": "Qwen/Qwen2.5-Coder-32B-Instruct", - "name": "Qwen/Qwen2.5-Coder-32B-Instruct", - "family": "qwen", - "attachment": false, + "claude-3-5-haiku-20241022": { + "id": "claude-3-5-haiku-20241022", + "name": "claude-3-5-haiku-20241022", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-11-11", - "last_updated": "2025-11-25", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 0.8, + "output": 4 } }, - "Qwen/Qwen2.5-7B-Instruct": { - "id": "Qwen/Qwen2.5-7B-Instruct", - "name": "Qwen/Qwen2.5-7B-Instruct", - "family": "qwen", + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM-4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 0.286, + "output": 1.142 } }, - "Qwen/Qwen3-VL-235B-A22B-Thinking": { - "id": "Qwen/Qwen3-VL-235B-A22B-Thinking", - "name": "Qwen/Qwen3-VL-235B-A22B-Thinking", - "family": "qwen", + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "gpt-5-pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-08", + "last_updated": "2025-10-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 400000, + "input": 272000, + "output": 272000 }, "cost": { - "input": 0.45, - "output": 3.5 + "input": 15, + "output": 120 } }, - "Qwen/Qwen3-30B-A3B-Thinking-2507": { - "id": "Qwen/Qwen3-30B-A3B-Thinking-2507", - "name": "Qwen/Qwen3-30B-A3B-Thinking-2507", - "family": "qwen", - "attachment": false, + "grok-4.20-beta-0309-reasoning": { + "id": "grok-4.20-beta-0309-reasoning", + "name": "grok-4.20-beta-0309-reasoning", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-31", - "last_updated": "2025-11-25", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 131000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.09, - "output": 0.3 + "input": 2, + "output": 6 } }, - "Qwen/Qwen3-VL-32B-Thinking": { - "id": "Qwen/Qwen3-VL-32B-Thinking", - "name": "Qwen/Qwen3-VL-32B-Thinking", - "family": "qwen", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "gemini-2.5-flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-21", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 0.3, + "output": 2.5 } }, - "Qwen/Qwen2.5-VL-72B-Instruct": { - "id": "Qwen/Qwen2.5-VL-72B-Instruct", - "name": "Qwen/Qwen2.5-VL-72B-Instruct", - "family": "qwen", + "gpt-4o": { + "id": "gpt-4o", + "name": "gpt-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-01-28", - "last_updated": "2025-11-25", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 4000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.59, - "output": 0.59 + "input": 2.5, + "output": 10 } }, - "stepfun-ai/Step-3.5-Flash": { - "id": "stepfun-ai/Step-3.5-Flash", - "name": "stepfun-ai/Step-3.5-Flash", - "family": "step", + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2025-12-19", + "last_updated": "2025-12-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.3, + "output": 1.2 } }, - "zai-org/GLM-4.5V": { - "id": "zai-org/GLM-4.5V", - "name": "zai-org/GLM-4.5V", - "family": "glm", + "gemini-2.5-flash-lite-preview-09-2025": { + "id": "gemini-2.5-flash-lite-preview-09-2025", + "name": "gemini-2.5-flash-lite-preview-09-2025", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2025-09-26", + "last_updated": "2025-09-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.14, - "output": 0.86 + "input": 0.1, + "output": 0.4 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "zai-org/GLM-4.6", - "family": "glm", - "attachment": false, + "doubao-seed-1-6-vision-250815": { + "id": "doubao-seed-1-6-vision-250815", + "name": "doubao-seed-1-6-vision-250815", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 256000, + "output": 32000 }, "cost": { - "input": 0.5, - "output": 1.9 + "input": 0.114, + "output": 1.143 } }, - "zai-org/GLM-4.6V": { - "id": "zai-org/GLM-4.6V", - "name": "zai-org/GLM-4.6V", - "family": "glm", + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "claude-opus-4-1-20250805", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-12-07", - "last_updated": "2025-12-07", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 15, + "output": 75 } }, - "zai-org/GLM-4.5-Air": { - "id": "zai-org/GLM-4.5-Air", - "name": "zai-org/GLM-4.5-Air", - "family": "glm-air", + "qwen3-max-2025-09-23": { + "id": "qwen3-max-2025-09-23", + "name": "qwen3-max-2025-09-23", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-09-24", + "last_updated": "2025-09-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 258048, + "output": 65536 }, "cost": { - "input": 0.14, - "output": 0.86 + "input": 0.86, + "output": 3.43 } }, - "inclusionAI/Ling-flash-2.0": { - "id": "inclusionAI/Ling-flash-2.0", - "name": "inclusionAI/Ling-flash-2.0", - "family": "ling", + "glm-4.7-flashx": { + "id": "glm-4.7-flashx", + "name": "glm-4.7-flashx", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2026-01-20", + "last_updated": "2026-01-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.0715, + "output": 0.429 } }, - "inclusionAI/Ling-mini-2.0": { - "id": "inclusionAI/Ling-mini-2.0", - "name": "inclusionAI/Ling-mini-2.0", - "family": "ling", + "glm-5.1": { + "id": "glm-5.1", + "name": "glm-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-09-10", - "last_updated": "2025-11-25", + "release_date": "2026-04-10", + "last_updated": "2026-04-10", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.86, + "output": 3.5 } }, - "inclusionAI/Ring-flash-2.0": { - "id": "inclusionAI/Ring-flash-2.0", - "name": "inclusionAI/Ring-flash-2.0", - "family": "ring", + "glm-4.6": { + "id": "glm-4.6", + "name": "glm-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.286, + "output": 1.142 } }, - "ascend-tribe/pangu-pro-moe": { - "id": "ascend-tribe/pangu-pro-moe", - "name": "ascend-tribe/pangu-pro-moe", - "family": "pangu", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "kimi-k2-thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", "attachment": false, "reasoning": true, - "tool_call": false, - "structured_output": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2025-07-02", - "last_updated": "2026-01-16", + "knowledge": "2025-06", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.575, + "output": 2.3 } }, - "tencent/Hunyuan-MT-7B": { - "id": "tencent/Hunyuan-MT-7B", - "name": "tencent/Hunyuan-MT-7B", - "family": "hunyuan", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "claude-sonnet-4-5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "knowledge": "2025-07-31", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 33000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15 } }, - "tencent/Hunyuan-A13B-Instruct": { - "id": "tencent/Hunyuan-A13B-Instruct", - "name": "tencent/Hunyuan-A13B-Instruct", - "family": "hunyuan", + "glm-4.5-x": { + "id": "glm-4.5-x", + "name": "glm-4.5-x", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 1.143, + "output": 2.29 } }, - "Pro/zai-org/GLM-4.7": { - "id": "Pro/zai-org/GLM-4.7", - "name": "Pro/zai-org/GLM-4.7", - "family": "glm", + "deepseek-v3.2-thinking": { + "id": "deepseek-v3.2-thinking", + "name": "DeepSeek-V3.2-Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-12", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.29, + "output": 0.43 } }, - "Pro/zai-org/GLM-5.1": { - "id": "Pro/zai-org/GLM-5.1", - "name": "Pro/zai-org/GLM-5.1", - "family": "glm", - "attachment": false, + "claude-sonnet-4-6-thinking": { + "id": "claude-sonnet-4-6-thinking", + "name": "claude-sonnet-4-6-thinking", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "knowledge": "2025-08", + "release_date": "2026-02-18", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_write": 0 + "input": 3, + "output": 15 } }, - "Pro/zai-org/GLM-5": { - "id": "Pro/zai-org/GLM-5", - "name": "Pro/zai-org/GLM-5", - "family": "glm", - "attachment": false, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "claude-opus-4-7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2026-01-31", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1, - "output": 3.2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "Pro/deepseek-ai/DeepSeek-V3": { - "id": "Pro/deepseek-ai/DeepSeek-V3", - "name": "Pro/deepseek-ai/DeepSeek-V3", - "family": "deepseek", - "attachment": false, + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "grok-4-1-fast-non-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-26", - "last_updated": "2025-11-25", + "knowledge": "2025-06", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.2, + "output": 0.5 } }, - "Pro/deepseek-ai/DeepSeek-R1": { - "id": "Pro/deepseek-ai/DeepSeek-R1", - "name": "Pro/deepseek-ai/DeepSeek-R1", - "family": "deepseek-thinking", - "attachment": false, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "gpt-5.4-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 2.18 + "input": 0.2, + "output": 1.25 } }, - "Pro/deepseek-ai/DeepSeek-V3.2": { - "id": "Pro/deepseek-ai/DeepSeek-V3.2", - "name": "Pro/deepseek-ai/DeepSeek-V3.2", - "family": "deepseek", - "attachment": false, + "claude-opus-4-1-20250805-thinking": { + "id": "claude-opus-4-1-20250805-thinking", + "name": "claude-opus-4-1-20250805-thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "knowledge": "2025-03", + "release_date": "2025-05-27", + "last_updated": "2025-05-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.27, - "output": 0.42 + "input": 15, + "output": 75 } }, - "Pro/deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "Pro/deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "Pro/deepseek-ai/DeepSeek-V3.1-Terminus", - "family": "deepseek", + "glm-4.5-airx": { + "id": "glm-4.5-airx", + "name": "glm-4.5-airx", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.27, - "output": 1 + "input": 0.572, + "output": 1.714 } }, - "Pro/moonshotai/Kimi-K2-Thinking": { - "id": "Pro/moonshotai/Kimi-K2-Thinking", - "name": "Pro/moonshotai/Kimi-K2-Thinking", - "family": "kimi-thinking", - "attachment": false, - "reasoning": true, + "grok-4.1": { + "id": "grok-4.1", + "name": "grok-4.1", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "attachment": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-11-07", - "last_updated": "2025-11-25", + "knowledge": "2025-06", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.55, - "output": 2.5 + "input": 2, + "output": 10 } }, - "Pro/moonshotai/Kimi-K2.6": { - "id": "Pro/moonshotai/Kimi-K2.6", - "name": "Pro/moonshotai/Kimi-K2.6", - "family": "kimi", - "attachment": false, - "reasoning": true, + "gemini-2.5-flash-preview-09-2025": { + "id": "gemini-2.5-flash-preview-09-2025", + "name": "gemini-2.5-flash-preview-09-2025", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-01", + "release_date": "2025-09-26", + "last_updated": "2025-09-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.3, + "output": 2.5 } }, - "Pro/moonshotai/Kimi-K2-Instruct-0905": { - "id": "Pro/moonshotai/Kimi-K2-Instruct-0905", - "name": "Pro/moonshotai/Kimi-K2-Instruct-0905", - "family": "kimi", - "attachment": false, - "reasoning": false, + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "claude-opus-4-5-20251101", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-08", + "knowledge": "2025-03-31", + "release_date": "2025-11-25", "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 5, + "output": 25 } }, - "Pro/moonshotai/Kimi-K2.5": { - "id": "Pro/moonshotai/Kimi-K2.5", - "name": "Pro/moonshotai/Kimi-K2.5", - "family": "kimi", - "attachment": false, + "claude-opus-4-20250514": { + "id": "claude-opus-4-20250514", + "name": "claude-opus-4-20250514", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.45, - "output": 2.25 + "input": 15, + "output": 75 } }, - "Pro/MiniMaxAI/MiniMax-M2.5": { - "id": "Pro/MiniMaxAI/MiniMax-M2.5", - "name": "Pro/MiniMaxAI/MiniMax-M2.5", - "family": "minimax", - "attachment": false, + "gemini-3-pro-image-preview": { + "id": "gemini-3-pro-image-preview", + "name": "gemini-3-pro-image-preview", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": false, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "knowledge": "2025-06", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 192000, - "output": 131000 + "context": 32768, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 1.22 + "input": 2, + "output": 120 } }, - "Pro/MiniMaxAI/MiniMax-M2.1": { - "id": "Pro/MiniMaxAI/MiniMax-M2.1", - "name": "Pro/MiniMaxAI/MiniMax-M2.1", - "family": "minimax", - "attachment": false, + "gemini-2.5-flash-image": { + "id": "gemini-2.5-flash-image", + "name": "gemini-2.5-flash-image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2025-01", + "release_date": "2025-10-08", + "last_updated": "2025-10-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 197000, - "output": 131000 + "context": 32768, + "output": 32768 }, "cost": { "input": 0.3, - "output": 1.2 + "output": 30 } }, - "PaddlePaddle/PaddleOCR-VL": { - "id": "PaddlePaddle/PaddleOCR-VL", - "name": "PaddlePaddle/PaddleOCR-VL", - "attachment": true, - "reasoning": false, - "tool_call": false, + "glm-for-coding": { + "id": "glm-for-coding", + "name": "glm-for-coding", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2025-10-16", - "last_updated": "2025-10-16", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.086, + "output": 0.343 } }, - "PaddlePaddle/PaddleOCR-VL-1.5": { - "id": "PaddlePaddle/PaddleOCR-VL-1.5", - "name": "PaddlePaddle/PaddleOCR-VL-1.5", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "gpt-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-12", + "last_updated": "2025-12-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.75, + "output": 14 } }, - "deepseek-ai/DeepSeek-OCR": { - "id": "deepseek-ai/DeepSeek-OCR", - "name": "deepseek-ai/DeepSeek-OCR", - "attachment": true, + "qwen-flash": { + "id": "qwen-flash", + "name": "Qwen-Flash", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-10-20", - "last_updated": "2025-10-20", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.022, + "output": 0.22 } }, - "deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "deepseek-ai/DeepSeek-V3.1-Terminus", - "family": "deepseek", - "attachment": false, + "claude-opus-4-6-thinking": { + "id": "claude-opus-4-6-thinking", + "name": "claude-opus-4-6-thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "knowledge": "2025-05", + "release_date": "2026-02-06", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 1 + "input": 5, + "output": 25 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "deepseek-ai/DeepSeek-V3.2", - "family": "deepseek", - "attachment": false, + "claude-sonnet-4-20250514": { + "id": "claude-sonnet-4-20250514", + "name": "claude-sonnet-4-20250514", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.27, - "output": 0.42 + "input": 3, + "output": 15 } }, - "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B": { - "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B", - "name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B", - "family": "qwen", - "attachment": false, + "gpt-5.1-chat-latest": { + "id": "gpt-5.1-chat-latest", + "name": "gpt-5.1-chat-latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.25, + "output": 10 } }, - "deepseek-ai/DeepSeek-R1": { - "id": "deepseek-ai/DeepSeek-R1", - "name": "deepseek-ai/DeepSeek-R1", - "family": "deepseek-thinking", - "attachment": false, + "gpt-5.2-chat-latest": { + "id": "gpt-5.2-chat-latest", + "name": "gpt-5.2-chat-latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-12", + "last_updated": "2025-12-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 2.18 + "input": 1.75, + "output": 14 } }, - "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B": { - "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", - "name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", - "family": "qwen", - "attachment": false, + "grok-4-fast-reasoning": { + "id": "grok-4-fast-reasoning", + "name": "grok-4-fast-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-11-25", + "knowledge": "2025-06", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 0.2, + "output": 0.5 } }, - "deepseek-ai/DeepSeek-V3": { - "id": "deepseek-ai/DeepSeek-V3", - "name": "deepseek-ai/DeepSeek-V3", - "family": "deepseek", - "attachment": false, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "gpt-4.1-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-12-26", - "last_updated": "2025-11-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.1, + "output": 0.4 } }, - "deepseek-ai/deepseek-vl2": { - "id": "deepseek-ai/deepseek-vl2", - "name": "deepseek-ai/deepseek-vl2", - "family": "deepseek", + "claude-sonnet-4-5-20250929-thinking": { + "id": "claude-sonnet-4-5-20250929-thinking", + "name": "claude-sonnet-4-5-20250929-thinking", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-13", - "last_updated": "2025-11-25", + "knowledge": "2025-03", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4000, - "output": 4000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 3, + "output": 15 } }, - "baidu/ERNIE-4.5-300B-A47B": { - "id": "baidu/ERNIE-4.5-300B-A47B", - "name": "baidu/ERNIE-4.5-300B-A47B", - "family": "ernie", + "MiniMax-M2.7-highspeed": { + "id": "MiniMax-M2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-02", - "last_updated": "2025-11-25", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 0.6, + "output": 4.8 } }, - "THUDM/GLM-Z1-32B-0414": { - "id": "THUDM/GLM-Z1-32B-0414", - "name": "THUDM/GLM-Z1-32B-0414", - "family": "glm-z", + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "release_date": "2025-10-26", + "last_updated": "2025-10-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.33, + "output": 1.32 } }, - "THUDM/GLM-4-32B-0414": { - "id": "THUDM/GLM-4-32B-0414", - "name": "THUDM/GLM-4-32B-0414", - "family": "glm", - "attachment": false, + "gemini-3.1-flash-image-preview": { + "id": "gemini-3.1-flash-image-preview", + "name": "gemini-3.1-flash-image-preview", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2026-02-27", + "last_updated": "2026-02-27", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 33000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.27, - "output": 0.27 + "input": 0.5, + "output": 60 } }, - "THUDM/GLM-4-9B-0414": { - "id": "THUDM/GLM-4-9B-0414", - "name": "THUDM/GLM-4-9B-0414", - "family": "glm", + "qwen3-235b-a22b": { + "id": "qwen3-235b-a22b", + "name": "Qwen3-235B-A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 33000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.086, - "output": 0.086 + "input": 0.29, + "output": 2.86 } }, - "THUDM/GLM-Z1-9B-0414": { - "id": "THUDM/GLM-Z1-9B-0414", - "name": "THUDM/GLM-Z1-9B-0414", - "family": "glm-z", - "attachment": false, - "reasoning": true, + "ministral-14b-2512": { + "id": "ministral-14b-2512", + "name": "ministral-14b-2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "attachment": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2024-12", + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.086, - "output": 0.086 + "input": 0.33, + "output": 0.33 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "moonshotai/Kimi-K2-Instruct-0905", - "family": "kimi", - "attachment": false, - "reasoning": false, + "glm-4.6v": { + "id": "glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-08", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.145, + "output": 0.43 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "moonshotai/Kimi-K2-Thinking", - "family": "kimi-thinking", - "attachment": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "claude-haiku-4-5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-11-07", - "last_updated": "2025-11-25", + "knowledge": "2025-02-28", + "release_date": "2025-10-16", + "last_updated": "2025-10-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.55, - "output": 2.5 + "input": 1, + "output": 5 } }, - "ByteDance-Seed/Seed-OSS-36B-Instruct": { - "id": "ByteDance-Seed/Seed-OSS-36B-Instruct", - "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", - "family": "seed", - "attachment": false, - "reasoning": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "gpt-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-09-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.21, - "output": 0.57 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 0, + "tiers": [ + { + "input": 5, + "output": 22.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5 + } } - } - } - }, - "submodel": { - "id": "submodel", - "env": ["SUBMODEL_INSTAGEN_ACCESS_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://llm.submodel.ai/v1", - "name": "submodel", - "doc": "https://submodel.gitbook.io", - "models": { - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + }, + "doubao-seed-1-6-thinking-250715": { + "id": "doubao-seed-1-6-thinking-250715", + "name": "doubao-seed-1-6-thinking-250715", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "release_date": "2025-07-15", + "last_updated": "2025-07-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 256000, + "output": 16000 }, "cost": { - "input": 0.2, - "output": 0.3 + "input": 0.121, + "output": 1.21 } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "gpt-5.4-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.75, + "output": 4.5 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3 235B A22B Thinking 2507", - "family": "qwen", - "attachment": false, - "reasoning": true, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "gpt-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 2, + "output": 8 } }, - "zai-org/GLM-4.5-Air": { - "id": "zai-org/GLM-4.5-Air", - "name": "GLM 4.5 Air", - "family": "glm-air", - "attachment": false, + "doubao-seed-code-preview-251028": { + "id": "doubao-seed-code-preview-251028", + "name": "doubao-seed-code-preview-251028", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2025-11-11", + "last_updated": "2025-11-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 256000, + "output": 32000 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.17, + "output": 1.14 } }, - "zai-org/GLM-4.5-FP8": { - "id": "zai-org/GLM-4.5-FP8", - "name": "GLM 4.5 FP8", - "family": "glm", - "attachment": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "claude-opus-4-6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-05-31", + "release_date": "2026-02-06", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 5, + "output": 25 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "qwen3-coder-480b-a35b-instruct": { + "id": "qwen3-coder-480b-a35b-instruct", + "name": "qwen3-coder-480b-a35b-instruct", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 75000, - "output": 163840 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.86, + "output": 3.43 } }, - "deepseek-ai/DeepSeek-V3-0324": { - "id": "deepseek-ai/DeepSeek-V3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "claude-sonnet-4-5-20250929", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "knowledge": "2025-07-31", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 75000, - "output": 163840 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 3, + "output": 15 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek R1 0528", + "deepseek-reasoner": { + "id": "deepseek-reasoner", + "name": "Deepseek-Reasoner", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 75000, - "output": 163840 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 2.15 + "input": 0.29, + "output": 0.43 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "grok-4-1-fast-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-08-23", - "last_updated": "2025-08-23", + "knowledge": "2025-06", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.1, + "input": 0.2, "output": 0.5 } - } - } - }, - "minimax-coding-plan": { - "id": "minimax-coding-plan", - "env": ["MINIMAX_API_KEY"], - "npm": "@ai-sdk/anthropic", - "api": "https://api.minimax.io/anthropic/v1", - "name": "MiniMax Coding Plan (minimax.io)", - "doc": "https://platform.minimax.io/docs/coding-plan/intro", - "models": { - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax-M2", - "family": "minimax", - "attachment": false, - "reasoning": true, + }, + "gemini-3-pro-preview": { + "id": "gemini-3-pro-preview", + "name": "gemini-3-pro-preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2025-06", + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 12 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, + "gpt-5-thinking": { + "id": "gpt-5-thinking", + "name": "gpt-5-thinking", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-10", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1.25, + "output": 10 } }, - "MiniMax-M2.7": { - "id": "MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "gpt-5-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2 + } + }, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "gpt-4.1-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.4, + "output": 1.6 } }, - "MiniMax-M2.7-highspeed": { - "id": "MiniMax-M2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", - "family": "minimax", - "attachment": false, + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, + "context": 200000, "output": 131072 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.72, + "output": 3.2 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", - "name": "MiniMax-M2.1", - "family": "minimax", - "attachment": false, + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "gpt-5.4-pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 30, + "output": 180, + "cache_read": 0, + "cache_write": 0, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } } }, - "MiniMax-M2.5-highspeed": { - "id": "MiniMax-M2.5-highspeed", - "name": "MiniMax-M2.5-highspeed", - "family": "minimax", + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "glm-4.5-air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.1143, + "output": 0.286 } - } - } - }, - "perplexity": { - "id": "perplexity", - "env": ["PERPLEXITY_API_KEY"], - "npm": "@ai-sdk/perplexity", - "name": "Perplexity", - "doc": "https://docs.perplexity.ai", - "models": { - "sonar-pro": { - "id": "sonar-pro", - "name": "Sonar Pro", - "family": "sonar-pro", + }, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "claude-sonnet-4-6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-18", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 1000000, + "output": 64000 }, "cost": { "input": 3, "output": 15 } }, - "sonar-deep-research": { - "id": "sonar-deep-research", - "name": "Perplexity Sonar Deep Research", - "attachment": false, - "reasoning": true, - "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-02-01", - "last_updated": "2025-09-01", + "grok-4-fast-non-reasoning": { + "id": "grok-4-fast-non-reasoning", + "name": "grok-4-fast-non-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 2, - "output": 8, - "reasoning": 3 + "input": 0.2, + "output": 0.5 } }, - "sonar": { - "id": "sonar", - "name": "Sonar", - "family": "sonar", - "attachment": false, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "gemini-3-flash-preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "knowledge": "2025-06", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 1, - "output": 1 + "input": 0.5, + "output": 3 } }, - "sonar-reasoning-pro": { - "id": "sonar-reasoning-pro", - "name": "Sonar Reasoning Pro", - "family": "sonar-reasoning", - "attachment": true, - "reasoning": true, - "tool_call": false, + "deepseek-chat": { + "id": "deepseek-chat", + "name": "Deepseek-Chat", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "knowledge": "2024-07", + "release_date": "2024-11-29", + "last_updated": "2024-11-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 8192 }, "cost": { - "input": 2, - "output": 8 + "input": 0.29, + "output": 0.43 } - } - } - }, - "deepseek": { - "id": "deepseek", - "env": ["DEEPSEEK_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.deepseek.com", - "name": "DeepSeek", - "doc": "https://api-docs.deepseek.com/quick_start/pricing", - "models": { - "deepseek-chat": { - "id": "deepseek-chat", - "name": "DeepSeek Chat", - "family": "deepseek", - "attachment": true, + }, + "MiniMax-M1": { + "id": "MiniMax-M1", + "name": "MiniMax-M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-12-01", - "last_updated": "2026-02-28", + "release_date": "2025-06-16", + "last_updated": "2025-06-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.132, + "output": 1.254 } }, - "deepseek-reasoner": { - "id": "deepseek-reasoner", - "name": "DeepSeek Reasoner", - "family": "deepseek-thinking", + "grok-4.20-multi-agent-beta-0309": { + "id": "grok-4.20-multi-agent-beta-0309", + "name": "grok-4.20-multi-agent-beta-0309", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-12-01", - "last_updated": "2026-02-28", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 2, + "output": 6 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "glm-5": { + "id": "glm-5", + "name": "glm-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.6, + "output": 2.6 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "qwen-max-latest": { + "id": "qwen-max-latest", + "name": "Qwen-Max-Latest", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2024-11", + "release_date": "2024-04-03", + "last_updated": "2025-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0.343, + "output": 1.372 } - } - } - }, - "llama": { - "id": "llama", - "env": ["LLAMA_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.llama.com/compat/v1/", - "name": "Llama", - "doc": "https://llama.developer.meta.com/docs/models", - "models": { - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", + }, + "mistral-large-2512": { + "id": "mistral-large-2512", + "name": "mistral-large-2512", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2024-12", + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 1.1, + "output": 3.3 } }, - "cerebras-llama-4-maverick-17b-128e-instruct": { - "id": "cerebras-llama-4-maverick-17b-128e-instruct", - "name": "Cerebras-Llama-4-Maverick-17B-128E-Instruct", - "family": "llama", - "attachment": true, + "MiniMax-M2.7": { + "id": "MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2 } }, - "llama-3.3-8b-instruct": { - "id": "llama-3.3-8b-instruct", - "name": "Llama-3.3-8B-Instruct", - "family": "llama", + "claude-3-5-haiku-latest": { + "id": "claude-3-5-haiku-latest", + "name": "claude-3-5-haiku-latest", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.8, + "output": 4 } }, - "cerebras-llama-4-scout-17b-16e-instruct": { - "id": "cerebras-llama-4-scout-17b-16e-instruct", - "name": "Cerebras-Llama-4-Scout-17B-16E-Instruct", - "family": "llama", - "attachment": true, + "qwen3-30b-a3b": { + "id": "qwen3-30b-a3b", + "name": "Qwen3-30B-A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2025-04", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.11, + "output": 1.08 } }, - "groq-llama-4-maverick-17b-128e-instruct": { - "id": "groq-llama-4-maverick-17b-128e-instruct", - "name": "Groq-Llama-4-Maverick-17B-128E-Instruct", - "family": "llama", - "attachment": true, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "deepseek-v3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2024-12", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.29, + "output": 0.43 } }, - "llama-4-scout-17b-16e-instruct-fp8": { - "id": "llama-4-scout-17b-16e-instruct-fp8", - "name": "Llama-4-Scout-17B-16E-Instruct-FP8", - "family": "llama", + "doubao-seed-1-8-251215": { + "id": "doubao-seed-1-8-251215", + "name": "doubao-seed-1-8-251215", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 224000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.114, + "output": 0.286 } }, - "llama-4-maverick-17b-128e-instruct-fp8": { - "id": "llama-4-maverick-17b-128e-instruct-fp8", - "name": "Llama-4-Maverick-17B-128E-Instruct-FP8", - "family": "llama", + "claude-opus-4-5-20251101-thinking": { + "id": "claude-opus-4-5-20251101-thinking", + "name": "claude-opus-4-5-20251101-thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2025-03", + "release_date": "2025-11-25", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25 } - } - } - }, - "openrouter": { - "id": "openrouter", - "env": ["OPENROUTER_API_KEY"], - "npm": "@openrouter/ai-sdk-provider", - "api": "https://openrouter.ai/api/v1", - "name": "OpenRouter", - "doc": "https://openrouter.ai/models", - "models": { - "liquid/lfm-2.5-1.2b-instruct:free": { - "id": "liquid/lfm-2.5-1.2b-instruct:free", - "name": "LFM2.5-1.2B-Instruct (free)", - "family": "liquid", - "attachment": false, + }, + "gemini-2.0-flash-lite": { + "id": "gemini-2.0-flash-lite", + "name": "gemini-2.0-flash-lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": false, "tool_call": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-20", - "last_updated": "2026-01-28", + "knowledge": "2024-11", + "release_date": "2025-06-16", + "last_updated": "2025-06-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 2000000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.075, + "output": 0.3 } }, - "liquid/lfm-2.5-1.2b-thinking:free": { - "id": "liquid/lfm-2.5-1.2b-thinking:free", - "name": "LFM2.5-1.2B-Thinking (free)", - "family": "liquid", + "glm-5-turbo": { + "id": "glm-5-turbo", + "name": "glm-5-turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-20", - "last_updated": "2026-01-28", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.72, + "output": 3.2 } }, - "deepseek/deepseek-chat-v3.1": { - "id": "deepseek/deepseek-chat-v3.1", - "name": "DeepSeek-V3.1", - "family": "deepseek", - "attachment": false, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "gpt-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 1.25, + "output": 10 } - }, - "deepseek/deepseek-r1-distill-llama-70b": { - "id": "deepseek/deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", - "attachment": false, + } + } + }, + "trustedrouter": { + "id": "trustedrouter", + "env": ["TRUSTEDROUTER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.trustedrouter.com/v1", + "name": "TrustedRouter", + "doc": "https://trustedrouter.com/docs", + "models": { + "zdr": { + "id": "zdr", + "name": "Zero Data Retention", + "description": "TrustedRouter privacy routing alias that prefers zero data retention model endpoints.", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-01-23", - "last_updated": "2025-01-23", + "release_date": "2026-06-01", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 - }, - "cost": { - "input": 0, - "output": 0 + "context": 1000000, + "output": 131072 } }, - "deepseek/deepseek-r1": { - "id": "deepseek/deepseek-r1", - "name": "DeepSeek: R1", - "family": "deepseek-thinking", - "attachment": false, + "e2e": { + "id": "e2e", + "name": "End-to-End Encrypted", + "description": "TrustedRouter privacy routing alias for end-to-end encrypted provider routes where available.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2026-06-01", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 16000 - }, - "cost": { - "input": 0.7, - "output": 2.5 + "context": 1000000, + "output": 131072 } }, - "deepseek/deepseek-v3.2-speciale": { - "id": "deepseek/deepseek-v3.2-speciale", - "name": "DeepSeek V3.2 Speciale", - "family": "deepseek", - "attachment": false, + "synth-code": { + "id": "synth-code", + "name": "Synth Code", + "description": "TrustedRouter code synthesis orchestration alias that combines multiple model responses into one answer.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2026-06-20", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 - }, - "cost": { - "input": 0.27, - "output": 0.41 + "context": 1000000, + "output": 131072 } }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, + "fast": { + "id": "fast", + "name": "Fast", + "description": "TrustedRouter speed routing alias that prefers low-latency healthy model endpoints.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2026-06-01", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 - }, - "cost": { - "input": 0.28, - "output": 0.4 + "context": 1000000, + "output": 131072 } }, - "deepseek/deepseek-v3.1-terminus:exacto": { - "id": "deepseek/deepseek-v3.1-terminus:exacto", - "name": "DeepSeek V3.1 Terminus (exacto)", - "family": "deepseek", - "attachment": false, + "synth": { + "id": "synth", + "name": "Synth", + "description": "TrustedRouter synthesis orchestration alias that combines multiple model responses into one answer.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "release_date": "2026-06-20", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 - }, - "cost": { - "input": 0.27, - "output": 1 + "context": 1000000, + "output": 131072 } }, - "deepseek/deepseek-chat-v3-0324": { - "id": "deepseek/deepseek-chat-v3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 16384, - "output": 8192 - }, - "cost": { - "input": 0, - "output": 0 - } - }, - "deepseek/deepseek-v3.1-terminus": { - "id": "deepseek/deepseek-v3.1-terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", - "attachment": false, + "auto": { + "id": "auto", + "name": "Auto", + "description": "TrustedRouter automatic routing alias that chooses a healthy supported model endpoint for the request.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "release_date": "2026-05-01", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 - }, - "cost": { - "input": 0.27, - "output": 1 + "context": 1000000, + "output": 131072 } }, - "openrouter/owl-alpha": { - "id": "openrouter/owl-alpha", - "name": "Owl Alpha", - "attachment": false, + "cheap": { + "id": "cheap", + "name": "Cheap", + "description": "TrustedRouter low-cost routing alias that prefers inexpensive healthy model endpoints.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-30", + "release_date": "2026-05-01", + "last_updated": "2026-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048756, - "output": 262144 - }, - "status": "alpha", - "cost": { - "input": 0, - "output": 0 + "context": 1000000, + "output": 131072 } - }, - "openrouter/pareto-code": { - "id": "openrouter/pareto-code", - "name": "Pareto Code Router", - "attachment": true, + } + } + }, + "zhipuai": { + "id": "zhipuai", + "env": ["ZHIPU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://open.bigmodel.cn/api/paas/v4", + "name": "Zhipu AI", + "doc": "https://docs.z.ai/guides/overview/pricing", + "models": { + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { "input": ["text"], "output": ["text"] @@ -22311,95 +25591,139 @@ "open_weights": false, "limit": { "context": 200000, - "output": 200000 + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 } }, - "openrouter/elephant-alpha": { - "id": "openrouter/elephant-alpha", - "name": "Elephant (free)", - "family": "elephant", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2026-04-13", - "last_updated": "2026-04-13", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 } }, - "openrouter/free": { - "id": "openrouter/free", - "name": "Free Models Router", + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "input": 200000, - "output": 8000 + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 22, + "cache_read": 1.2, + "cache_write": 0 } }, - "arcee-ai/trinity-large-thinking": { - "id": "arcee-ai/trinity-large-thinking", - "name": "Trinity Large Thinking", - "family": "trinity", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-03", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 80000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.22, - "output": 0.85 + "input": 1, + "output": 3.2, + "cache_read": 0.2, + "cache_write": 0 } }, - "arcee-ai/trinity-large-preview:free": { - "id": "arcee-ai/trinity-large-preview:free", - "name": "Trinity Large Preview", - "family": "trinity", + "glm-4.5-flash": { + "id": "glm-4.5-flash", + "name": "GLM-4.5-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-28", - "last_updated": "2026-01-28", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] @@ -22407,175 +25731,195 @@ "open_weights": true, "limit": { "context": 131072, - "output": 131072 + "output": 98304 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "cognitivecomputations/dolphin-mistral-24b-venice-edition:free": { - "id": "cognitivecomputations/dolphin-mistral-24b-venice-edition:free", - "name": "Uncensored (free)", - "family": "mistral", + "glm-4.7-flash": { + "id": "glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-07-09", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 200000, + "output": 131072 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "bytedance-seed/seedream-4.5": { - "id": "bytedance-seed/seedream-4.5", - "name": "Seedream 4.5", - "family": "seed", + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-23", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["image", "text"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 4096, - "output": 4096 - }, - "cost": { - "input": 0, - "output": 0 - } - }, - "black-forest-labs/flux.2-max": { - "id": "black-forest-labs/flux.2-max", - "name": "FLUX.2 Max", - "family": "flux", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-16", - "last_updated": "2026-01-31", - "modalities": { - "input": ["image", "text"], - "output": ["image"] - }, - "open_weights": false, - "limit": { - "context": 46864, - "output": 46864 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 1.1, + "cache_read": 0.03, + "cache_write": 0 } }, - "black-forest-labs/flux.2-flex": { - "id": "black-forest-labs/flux.2-flex", - "name": "FLUX.2 Flex", - "family": "flux", - "attachment": false, - "reasoning": false, - "tool_call": false, + "glm-4.6v": { + "id": "glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-25", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["image", "text"], - "output": ["image"] + "input": ["text", "image", "video"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 67344, - "output": 67344 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 0.9 } }, - "black-forest-labs/flux.2-pro": { - "id": "black-forest-labs/flux.2-pro", - "name": "FLUX.2 Pro", - "family": "flux", + "glm-4.6": { + "id": "glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-11-25", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["image", "text"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 46864, - "output": 46864 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 } }, - "black-forest-labs/flux.2-klein-4b": { - "id": "black-forest-labs/flux.2-klein-4b", - "name": "FLUX.2 Klein 4B", - "family": "flux", + "glm-4.7-flashx": { + "id": "glm-4.7-flashx", + "name": "GLM-4.7-FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-14", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["image", "text"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 40960 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.07, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0 } }, - "nousresearch/hermes-3-llama-3.1-405b:free": { - "id": "nousresearch/hermes-3-llama-3.1-405b:free", - "name": "Hermes 3 405B Instruct (free)", - "family": "hermes", + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-08-16", - "last_updated": "2024-08-16", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] @@ -22583,911 +25927,974 @@ "open_weights": true, "limit": { "context": 131072, - "output": 131072 + "output": 98304 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 } }, - "nousresearch/hermes-4-405b": { - "id": "nousresearch/hermes-4-405b", - "name": "Hermes 4 405B", - "family": "hermes", - "attachment": false, + "glm-4.5v": { + "id": "glm-4.5v", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-08-25", - "last_updated": "2025-08-25", + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 64000, + "output": 16384 }, "cost": { - "input": 1, - "output": 3 + "input": 0.6, + "output": 1.8 } }, - "nousresearch/hermes-4-70b": { - "id": "nousresearch/hermes-4-70b", - "name": "Hermes 4 70B", - "family": "hermes", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-08-25", - "last_updated": "2025-08-25", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 204800, "output": 131072 }, "cost": { - "input": 0.13, - "output": 0.4 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 } - }, - "stepfun/step-3.5-flash": { - "id": "stepfun/step-3.5-flash", - "name": "Step 3.5 Flash", - "family": "step", + } + } + }, + "cortecs": { + "id": "cortecs", + "env": ["CORTECS_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.cortecs.ai/v1", + "name": "Cortecs", + "doc": "https://api.cortecs.ai/v1/models", + "models": { + "deepseek-r1-0528": { + "id": "deepseek-r1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 164000, + "output": 164000 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.02 + "input": 0.585, + "output": 2.307 } }, - "mistralai/mistral-small-3.1-24b-instruct": { - "id": "mistralai/mistral-small-3.1-24b-instruct", - "name": "Mistral Small 3.1 24B Instruct", - "family": "mistral-small", - "attachment": true, - "reasoning": false, + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-17", - "last_updated": "2025-03-17", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.133, + "output": 0.266, + "cache_read": 0.0028 } }, - "mistralai/devstral-2512": { - "id": "mistralai/devstral-2512", - "name": "Devstral 2 2512", - "family": "devstral", + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-09-12", - "last_updated": "2025-09-12", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 196608, + "output": 196608 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.32, + "output": 1.18 } }, - "mistralai/codestral-2508": { - "id": "mistralai/codestral-2508", - "name": "Codestral 2508", - "family": "codestral", + "deepseek-v3-0324": { + "id": "deepseek-v3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "knowledge": "2024-07", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.551, + "output": 1.654 } }, - "mistralai/mistral-medium-3.1": { - "id": "mistralai/mistral-medium-3.1", - "name": "Mistral Medium 3.1", - "family": "mistral-medium", + "claude-opus4-7": { + "id": "claude-opus4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 5.6, + "output": 27.99, + "cache_read": 0.56, + "cache_write": 6.99 } }, - "mistralai/mistral-small-2603": { - "id": "mistralai/mistral-small-2603", - "name": "Mistral Small 4", - "family": "mistral-small", - "attachment": true, + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 198000, + "output": 198000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.45, + "output": 2.23 } }, - "mistralai/mistral-medium-3": { - "id": "mistralai/mistral-medium-3", - "name": "Mistral Medium 3", - "family": "mistral-medium", - "attachment": true, - "reasoning": false, + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 131000, + "output": 131000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.062, + "output": 0.408 } }, - "mistralai/devstral-small-2505": { - "id": "mistralai/devstral-small-2505", - "name": "Devstral Small", - "family": "devstral", + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "knowledge": "2025-04", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.06, - "output": 0.12 + "input": 0.053, + "output": 0.222 } }, - "mistralai/mistral-small-3.2-24b-instruct": { - "id": "mistralai/mistral-small-3.2-24b-instruct", - "name": "Mistral Small 3.2 24B Instruct", - "family": "mistral-small", - "attachment": true, - "reasoning": false, + "minimax-m2.1": { + "id": "minimax-m2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-06-20", - "last_updated": "2025-06-20", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 96000, - "output": 8192 + "context": 196000, + "output": 196000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.34, + "output": 1.34 } }, - "mistralai/devstral-medium-2507": { - "id": "mistralai/devstral-medium-2507", - "name": "Devstral Medium", - "family": "devstral", + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "knowledge": "2024-12", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.099, + "output": 0.33 } }, - "mistralai/devstral-small-2507": { - "id": "mistralai/devstral-small-2507", - "name": "Devstral Small 1.1", - "family": "devstral", - "attachment": false, - "reasoning": false, + "claude-4-6-sonnet": { + "id": "claude-4-6-sonnet", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 1000000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 3.59, + "output": 17.92 } }, - "meta-llama/llama-3.2-11b-vision-instruct": { - "id": "meta-llama/llama-3.2-11b-vision-instruct", - "name": "Llama 3.2 11B Vision Instruct", - "family": "llama", - "attachment": true, + "claude-sonnet-4": { + "id": "claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2025-03", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3.307, + "output": 16.536 } }, - "meta-llama/llama-3.2-3b-instruct:free": { - "id": "meta-llama/llama-3.2-3b-instruct:free", - "name": "Llama 3.2 3B Instruct (free)", + "llama-4-maverick": { + "id": "llama-4-maverick", + "name": "Llama 4 Maverick 17B Instruct", + "description": "Open multimodal Llama for strong reasoning with efficient everyday serving", "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.124, + "output": 0.603, + "cache_read": 0.03, + "cache_write": 0.151 } }, - "meta-llama/llama-3.3-70b-instruct:free": { - "id": "meta-llama/llama-3.3-70b-instruct:free", - "name": "Llama 3.3 70B Instruct (free)", - "family": "llama", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 1.654, + "output": 11.024 } }, - "x-ai/grok-4.20-multi-agent-beta": { - "id": "x-ai/grok-4.20-multi-agent-beta", - "name": "Grok 4.20 Multi - Agent Beta", - "family": "grok", + "claude-4-5-sonnet": { + "id": "claude-4-5-sonnet", + "name": "Claude 4.5 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2026-03-12", - "last_updated": "2026-03-12", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "output": 200000 }, - "status": "beta", "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - } + "input": 3.259, + "output": 16.296 } }, - "x-ai/grok-4-fast": { - "id": "x-ai/grok-4-fast", - "name": "Grok 4 Fast", - "family": "grok", + "mixtral-8x7B-instruct-v0.1": { + "id": "mixtral-8x7B-instruct-v0.1", + "name": "Mixtral 8x7B Instruct v0.1", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "knowledge": "2023-09", + "release_date": "2023-12-11", + "last_updated": "2023-12-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05, - "cache_write": 0.05 + "input": 0.438, + "output": 0.68 } }, - "x-ai/grok-code-fast-1": { - "id": "x-ai/grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM 4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 10000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.67, + "output": 2.46 } }, - "x-ai/grok-3-beta": { - "id": "x-ai/grok-3-beta", - "name": "Grok 3 Beta", - "family": "grok", + "kimi-k2-instruct": { + "id": "kimi-k2-instruct", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2024-07", + "release_date": "2025-07-11", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 131000, + "output": 131000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75, - "cache_write": 15 + "input": 0.551, + "output": 2.646 } }, - "x-ai/grok-4": { - "id": "x-ai/grok-4", - "name": "Grok 4", - "family": "grok", + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 131000, + "output": 131000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75, - "cache_write": 15, - "tiers": [ - { - "input": 6, - "output": 30, - "tier": { - "type": "context", - "size": 128000 - } - } - ] + "input": 0.089, + "output": 0.275 } }, - "x-ai/grok-3-mini": { - "id": "x-ai/grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", - "attachment": false, + "claude-opus4-8": { + "id": "claude-opus4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075, - "cache_write": 0.5 + "input": 5.64, + "output": 28.198, + "cache_read": 0.563, + "cache_write": 7.049 } }, - "x-ai/grok-4.1-fast": { - "id": "x-ai/grok-4.1-fast", - "name": "Grok 4.1 Fast", - "family": "grok", - "attachment": false, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 2000000, - "output": 30000 - }, - "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05, - "cache_write": 0.05 - } - }, - "x-ai/grok-4.20-beta": { - "id": "x-ai/grok-4.20-beta", - "name": "Grok 4.20 Beta", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-12", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "output": 262144 }, - "status": "beta", "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - } + "input": 1.28, + "output": 4.63, + "cache_read": 0.32 } }, - "x-ai/grok-3-mini-beta": { - "id": "x-ai/grok-3-mini-beta", - "name": "Grok 3 Mini Beta", - "family": "grok", + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "release_date": "2026-04-14", + "last_updated": "2026-04-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075, - "cache_write": 0.5 + "input": 1.31, + "output": 4.1, + "cache_read": 0.24 } }, - "x-ai/grok-3": { - "id": "x-ai/grok-3", - "name": "Grok 3", - "family": "grok", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75, - "cache_write": 15 + "input": 1.553, + "output": 3.106, + "cache_read": 0.003625 } }, - "tencent/hy3-preview": { - "id": "tencent/hy3-preview", - "name": "Hy3 preview", - "family": "Hy", + "qwen3-next-80b-a3b-thinking": { + "id": "qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B Thinking", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "knowledge": "2025-04", + "release_date": "2025-09-11", + "last_updated": "2025-09-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.066, - "output": 0.26, - "cache_read": 0.029, - "cache_write": 0.029 + "input": 0.164, + "output": 1.311 } }, - "poolside/laguna-m.1:free": { - "id": "poolside/laguna-m.1:free", - "name": "Laguna M.1", - "attachment": false, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-28", + "knowledge": "2025-12", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.656, + "output": 2.731 } }, - "poolside/laguna-xs.2:free": { - "id": "poolside/laguna-xs.2:free", - "name": "Laguna XS.2", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-28", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1.44, + "output": 4.53, + "cache_read": 0.39 } }, - "prime-intellect/intellect-3": { - "id": "prime-intellect/intellect-3", - "name": "Intellect 3", - "family": "glm", - "attachment": false, + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-01-15", - "last_updated": "2025-01-15", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 512000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 1.1 + "input": 0.355, + "output": 1.775, + "cache_read": 0.089 } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "Nemotron 3 Super", - "family": "nemotron", + "minimax-m2": { + "id": "minimax-m2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2024-11", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "output": 400000 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.39, + "output": 1.57 } }, - "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": { - "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free", - "name": "Nemotron 3 Nano Omni (free)", - "family": "nemotron", - "attachment": true, + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax-m2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-28", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 65536 + "context": 202752, + "output": 196072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.47, + "output": 1.4 } }, - "nvidia/nemotron-3-nano-30b-a3b:free": { - "id": "nvidia/nemotron-3-nano-30b-a3b:free", - "name": "Nemotron 3 Nano 30B A3B (free)", - "family": "nemotron", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-12-14", - "last_updated": "2026-01-31", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, @@ -23496,22 +26903,28 @@ "output": 256000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.55, + "output": 2.76 } }, - "nvidia/nemotron-nano-9b-v2:free": { - "id": "nvidia/nemotron-nano-9b-v2:free", - "name": "Nemotron Nano 9B V2 (free)", - "family": "nemotron", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT Oss 120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-09-05", - "last_updated": "2025-08-18", + "knowledge": "2024-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] @@ -23526,962 +26939,1037 @@ "output": 0 } }, - "nvidia/nemotron-3-super-120b-a12b:free": { - "id": "nvidia/nemotron-3-super-120b-a12b:free", - "name": "Nemotron 3 Super (free)", - "family": "nemotron", + "qwen-2.5-72b-instruct": { + "id": "qwen-2.5-72b-instruct", + "name": "Qwen2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2024-06", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 33000, + "output": 33000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.062, + "output": 0.231 } }, - "nvidia/nemotron-nano-9b-v2": { - "id": "nvidia/nemotron-nano-9b-v2", - "name": "nvidia-nemotron-nano-9b-v2", - "family": "nemotron", + "hermes-4-70b": { + "id": "hermes-4-70b", + "name": "Hermes 4 70B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-08-18", - "last_updated": "2025-08-18", + "knowledge": "2023-12", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.04, - "output": 0.16 + "input": 0.116, + "output": 0.358 } }, - "nvidia/nemotron-nano-12b-v2-vl:free": { - "id": "nvidia/nemotron-nano-12b-v2-vl:free", - "name": "Nemotron Nano 12B 2 VL (free)", - "family": "nemotron", - "attachment": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-10-28", - "last_updated": "2026-01-31", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 200000, + "output": 200000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.09, + "output": 5.43 } }, - "inception/mercury-edit-2": { - "id": "inception/mercury-edit-2", - "name": "Mercury Edit 2", - "attachment": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1050000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 3, + "output": 16.13, + "cache_read": 0.25 } }, - "inception/mercury-2": { - "id": "inception/mercury-2", - "name": "Mercury 2", - "family": "mercury", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-04", - "last_updated": "2026-03-04", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 50000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 0.81, + "output": 3.54, + "cache_read": 0.2 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "GPT-5.1-Codex-Max", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT 4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2024-06", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 2.354, + "output": 9.417 } }, - "openai/gpt-5.2-chat": { - "id": "openai/gpt-5.2-chat", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "nova-pro-v1": { + "id": "nova-pro-v1", + "name": "Nova Pro 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova-pro", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 300000, + "output": 5000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.016, + "output": 4.061 } }, - "openai/gpt-oss-120b:exacto": { - "id": "openai/gpt-oss-120b:exacto", - "name": "GPT OSS 120B (exacto)", - "family": "gpt-oss", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "Qwen3 Coder Next 80B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-04", + "release_date": "2026-02-04", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.05, - "output": 0.24 + "input": 0.158, + "output": 0.84 } }, - "openai/gpt-5-chat": { - "id": "openai/gpt-5-chat", - "name": "GPT-5 Chat (latest)", - "family": "gpt-codex", + "claude-opus4-5": { + "id": "claude-opus4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 200000 }, "cost": { - "input": 1.25, - "output": 10 + "input": 5.98, + "output": 29.89 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT-5.2 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "qwen3-coder-480b-a35b-instruct": { + "id": "qwen3-coder-480b-a35b-instruct", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 21, - "output": 168 + "input": 0.441, + "output": 1.984 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "devstral-2512": { + "id": "devstral-2512", + "name": "Devstral 2 2512", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.25, - "output": 2 + "input": 0, + "output": 0 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2026-01", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 250000, + "output": 250000 }, "cost": { - "input": 0.05, - "output": 0.4 + "input": 0.6, + "output": 3.6 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT-5.3-Codex", - "family": "gpt-codex", + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.235, + "output": 4.118, + "cache_read": 0.308, + "cache_write": 1.544 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-01", + "last_updated": "2025-08-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.22, + "output": 1.34 } }, - "openai/gpt-oss-20b:free": { - "id": "openai/gpt-oss-20b:free", - "name": "gpt-oss-20b (free)", - "family": "gpt-oss", + "glm-4.7-flash": { + "id": "glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2026-01-31", + "knowledge": "2025-04", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 203000, + "output": 203000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.09, + "output": 0.53 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o-mini", - "family": "gpt-mini", + "intellect-3": { + "id": "intellect-3", + "name": "INTELLECT 3", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2025-11", + "release_date": "2025-11-26", + "last_updated": "2025-11-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 0.219, + "output": 1.202 } }, - "openai/gpt-5.4-mini": { - "id": "openai/gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "llama-3.1-405b-instruct": { + "id": "llama-3.1-405b-instruct", + "name": "Llama 3.1 405B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, + "context": 128000, "output": 128000 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0, + "output": 0 } }, - "openai/gpt-5.1-chat": { - "id": "openai/gpt-5.1-chat", - "name": "GPT-5.1 Chat", - "family": "gpt-codex", - "attachment": true, + "glm-5": { + "id": "glm-5", + "name": "GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 202752, + "output": 202752 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1.08, + "output": 3.44 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "o4 Mini", - "family": "o-mini", + "mistral-large-2512": { + "id": "mistral-large-2512", + "name": "Mistral Large 3 2512", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "knowledge": "2025-12", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 0.5, + "output": 1.5, + "cache_read": 0.05 } }, - "openai/gpt-5.4-nano": { - "id": "openai/gpt-5.4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": false, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.266, + "output": 0.444 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "codestral-2508": { + "id": "codestral-2508", + "name": "Codestral 2508", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "mistral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "knowledge": "2025-03", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.3, + "output": 0.9, + "cache_read": 0.03 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT-5.1-Codex-Mini", - "family": "gpt-codex", - "attachment": true, + "qwen3.5-122b-a10b": { + "id": "qwen3.5-122b-a10b", + "name": "Qwen3.5 122B A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2026-01", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 100000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.444, + "output": 3.106 } }, - "openai/gpt-5-image": { - "id": "openai/gpt-5-image", - "name": "GPT-5 Image", - "family": "gpt", - "attachment": true, + "nemotron-3-super-120b-a12b": { + "id": "nemotron-3-super-120b-a12b", + "name": "Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-10-14", - "last_updated": "2025-10-14", + "knowledge": "2025-12", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 5, - "output": 10, - "cache_read": 1.25 + "input": 0.266, + "output": 0.799 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, + "glm-5-turbo": { + "id": "glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1.235, + "output": 4.118, + "cache_read": 0.308, + "cache_write": 1.544 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", + "claude-opus4-6": { + "id": "claude-opus4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1000000, + "output": 1000000 }, "cost": { - "input": 30, - "output": 180, - "cache_read": 30 + "input": 5.98, + "output": 29.89 } - }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5 Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + } + } + }, + "nebius": { + "id": "nebius", + "env": ["NEBIUS_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.tokenfactory.nebius.com/v1", + "name": "Nebius Token Factory", + "doc": "https://docs.tokenfactory.nebius.com/", + "models": { + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 - } - }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-08", + "release_date": "2025-12-05", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "input": 120000, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.2 + "input": 0.13, + "output": 0.4, + "cache_read": 0.013, + "cache_write": 0.16 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", + "moonshotai/Kimi-K2.5-fast": { + "id": "moonshotai/Kimi-K2.5-fast", + "name": "Kimi-K2.5-fast", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 272000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 15, - "output": 120 - } - }, - "openai/gpt-oss-120b:free": { - "id": "openai/gpt-oss-120b:free", - "name": "gpt-oss-120b (free)", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, - "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-06", + "release_date": "2025-12-15", + "last_updated": "2026-02-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 256000, + "input": 256000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 2.5, + "cache_read": 0.05, + "cache_write": 0.625 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", - "family": "gpt", + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi-K2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-06", + "release_date": "2025-12-15", + "last_updated": "2026-02-04", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 1.25, - "output": 10 + "input": 0.5, + "output": 2.5, + "reasoning": 2.5, + "cache_read": 0.05, + "cache_write": 0.625 } }, - "openai/gpt-oss-safeguard-20b": { - "id": "openai/gpt-oss-safeguard-20b", - "name": "GPT OSS Safeguard 20B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, + "google/gemma-3-27b-it": { + "id": "google/gemma-3-27b-it", + "name": "Gemma-3-27b-it", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-10-29", - "last_updated": "2025-10-29", + "knowledge": "2025-10", + "release_date": "2026-01-20", + "last_updated": "2026-02-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 110000, + "input": 100000, + "output": 8192 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01, + "cache_write": 0.125 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "Qwen/Qwen3-Next-80B-A3B-Thinking-fast": { + "id": "Qwen/Qwen3-Next-80B-A3B-Thinking-fast", + "name": "Qwen3-Next-80B-A3B-Thinking-fast", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-07", + "release_date": "2025-07-25", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.072, - "output": 0.28 + "input": 0.15, + "output": 1.2, + "cache_read": 0.015, + "cache_write": 0.1875 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "Qwen/Qwen2.5-VL-72B-Instruct": { + "id": "Qwen/Qwen2.5-VL-72B-Instruct", + "name": "Qwen2.5-VL-72B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2024-12", + "release_date": "2025-01-20", + "last_updated": "2026-02-04", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 128000, + "input": 120000, + "output": 8192 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.25, + "output": 0.75, + "cache_read": 0.025, + "cache_write": 0.31 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT-4.1 Mini", - "family": "gpt-mini", - "attachment": true, + "Qwen/Qwen3-Embedding-8B": { + "id": "Qwen/Qwen3-Embedding-8B", + "name": "Qwen3-Embedding-8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "temperature": false, + "knowledge": "2025-10", + "release_date": "2026-01-10", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 32768, + "input": 32768, + "output": 0 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.01, + "output": 0 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1-Codex", - "family": "gpt-codex", - "attachment": true, + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen3.5-397B-A17B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-07", + "release_date": "2025-07-15", + "last_updated": "2026-05-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "input": 250000, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.6, + "output": 3.6, + "cache_read": 0.06, + "cache_write": 0.75 } }, - "z-ai/glm-4.7": { - "id": "z-ai/glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "Qwen/Qwen3.5-397B-A17B-fast": { + "id": "Qwen/Qwen3.5-397B-A17B-fast", + "name": "Qwen3.5-397B-A17B-fast", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2025-07", + "release_date": "2025-07-15", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { "input": 0.6, - "output": 2.2, - "cache_read": 0.11 + "output": 3.6, + "cache_read": 0.06, + "cache_write": 0.75 } }, - "z-ai/glm-4.5-air:free": { - "id": "z-ai/glm-4.5-air:free", - "name": "GLM 4.5 Air (free)", - "family": "glm-air", + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen3-30B-A3B-Instruct-2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-12", + "release_date": "2026-01-28", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] @@ -24489,83 +27977,88 @@ "open_weights": true, "limit": { "context": 128000, - "output": 96000 + "input": 120000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01, + "cache_write": 0.125 } }, - "z-ai/glm-5": { - "id": "z-ai/glm-5", - "name": "GLM-5", - "family": "glm", + "Qwen/Qwen3-Next-80B-A3B-Thinking": { + "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", + "name": "Qwen3-Next-80B-A3B-Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-12", + "release_date": "2026-01-28", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131000 + "context": 128000, + "input": 120000, + "output": 16384 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.15, + "output": 1.2, + "reasoning": 1.2, + "cache_read": 0.015, + "cache_write": 0.18 } }, - "z-ai/glm-5.1": { - "id": "z-ai/glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "knowledge": "2025-07", + "release_date": "2025-07-25", + "last_updated": "2025-10-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 262144, + "output": 8192 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.2, + "output": 0.6 } }, - "z-ai/glm-4.5": { - "id": "z-ai/glm-4.5", - "name": "GLM 4.5", - "family": "glm", + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-12", + "release_date": "2026-01-28", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] @@ -24573,746 +28066,1071 @@ "open_weights": true, "limit": { "context": 128000, - "output": 96000 + "input": 120000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01, + "cache_write": 0.125 } }, - "z-ai/glm-4.6:exacto": { - "id": "z-ai/glm-4.6:exacto", - "name": "GLM 4.6 (exacto)", - "family": "glm", + "Qwen/Qwen3-235B-A22B-Thinking-2507-fast": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507-fast", + "name": "Qwen3-235B-A22B-Thinking-2507-fast", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-07", + "release_date": "2025-07-25", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.6, - "output": 1.9, - "cache_read": 0.11 + "input": 0.5, + "output": 2, + "cache_read": 0.05, + "cache_write": 0.625 } }, - "z-ai/glm-4.5-air": { - "id": "z-ai/glm-4.5-air", - "name": "GLM 4.5 Air", - "family": "glm-air", + "openai/gpt-oss-120b-fast": { + "id": "openai/gpt-oss-120b-fast", + "name": "gpt-oss-120b-fast", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-06", + "release_date": "2025-06-10", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 96000 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.2, - "output": 1.1 + "input": 0.1, + "output": 0.5, + "cache_read": 0.01, + "cache_write": 0.125 } }, - "z-ai/glm-5-turbo": { - "id": "z-ai/glm-5-turbo", - "name": "GLM-5-Turbo", - "family": "glm", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-09", + "release_date": "2026-01-10", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 128000, + "input": 124000, + "output": 8192 }, "cost": { - "input": 0.96, - "output": 3.2, - "cache_read": 0.192, - "cache_write": 0 + "input": 0.15, + "output": 0.6, + "reasoning": 0.6, + "cache_read": 0.015, + "cache_write": 0.18 } }, - "z-ai/glm-4.5v": { - "id": "z-ai/glm-4.5v", - "name": "GLM 4.5V", - "family": "glm", - "attachment": true, + "NousResearch/Hermes-4-405B": { + "id": "NousResearch/Hermes-4-405B", + "name": "Hermes-4-405B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "knowledge": "2025-11", + "release_date": "2026-01-30", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 64000, - "output": 16384 + "context": 128000, + "input": 120000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 1, + "output": 3, + "reasoning": 3, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "z-ai/glm-4.6": { - "id": "z-ai/glm-4.6", - "name": "GLM 4.6", - "family": "glm", + "NousResearch/Hermes-4-70B": { + "id": "NousResearch/Hermes-4-70B", + "name": "Hermes-4-70B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-11", + "release_date": "2026-01-30", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "input": 120000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11 + "input": 0.13, + "output": 0.4, + "reasoning": 0.4, + "cache_read": 0.013, + "cache_write": 0.16 } }, - "z-ai/glm-4.7-flash": { - "id": "z-ai/glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm", + "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B": { + "id": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B", + "name": "Nemotron-3-Nano-30B-A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, "structured_output": true, "temperature": true, - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "knowledge": "2025-05", + "release_date": "2025-08-10", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 65535 + "context": 32000, + "input": 30000, + "output": 4096 }, "cost": { - "input": 0.07, - "output": 0.4 + "input": 0.06, + "output": 0.24, + "cache_read": 0.006, + "cache_write": 0.075 } }, - "sourceful/riverflow-v2-standard-preview": { - "id": "sourceful/riverflow-v2-standard-preview", - "name": "Riverflow V2 Standard Preview", - "family": "sourceful", + "nvidia/Nemotron-3-Nano-Omni": { + "id": "nvidia/Nemotron-3-Nano-Omni", + "name": "Nemotron-3-Nano-Omni", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-08", - "last_updated": "2026-01-28", + "knowledge": "2025-01", + "release_date": "2025-01-20", + "last_updated": "2026-05-07", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, + "context": 65536, + "input": 60000, "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.06, + "output": 0.24, + "cache_read": 0.006, + "cache_write": 0.075 } }, - "sourceful/riverflow-v2-fast-preview": { - "id": "sourceful/riverflow-v2-fast-preview", - "name": "Riverflow V2 Fast Preview", - "family": "sourceful", + "nvidia/Llama-3_1-Nemotron-Ultra-253B-v1": { + "id": "nvidia/Llama-3_1-Nemotron-Ultra-253B-v1", + "name": "Llama-3.1-Nemotron-Ultra-253B-v1", + "description": "Flagship Nemotron model for high-throughput reasoning and complex agents", + "family": "nemotron", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-08", - "last_updated": "2026-01-28", + "knowledge": "2024-12", + "release_date": "2025-01-15", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 128000, + "input": 120000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 1.8, + "cache_read": 0.06, + "cache_write": 0.75 } }, - "sourceful/riverflow-v2-max-preview": { - "id": "sourceful/riverflow-v2-max-preview", - "name": "Riverflow V2 Max Preview", - "family": "sourceful", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "Nemotron-3-Super-120B-A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-08", - "last_updated": "2026-01-28", + "knowledge": "2026-02", + "release_date": "2026-03-11", + "last_updated": "2026-03-12", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 0.9 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "MiniMax M2.7", - "family": "minimax", + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2026-01", + "release_date": "2026-03-01", + "last_updated": "2026-03-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "input": 200000, + "output": 16384 }, + "status": "deprecated", "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 1, + "output": 3.2, + "cache_read": 0.1, + "cache_write": 1 } }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "MiniMax M2", - "family": "minimax", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { - "field": "reasoning_details" + "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2025-10-23", - "last_updated": "2025-10-23", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196600, - "output": 118000 + "context": 432000, + "output": 432000 }, "cost": { - "input": 0.28, - "output": 1.15, - "cache_read": 0.28, - "cache_write": 1.15 + "input": 1.4, + "output": 4.4 } }, - "minimax/minimax-01": { - "id": "minimax/minimax-01", - "name": "MiniMax-01", - "family": "minimax", - "attachment": true, + "deepseek-ai/DeepSeek-V3.2-fast": { + "id": "deepseek-ai/DeepSeek-V3.2-fast", + "name": "DeepSeek-V3.2-fast", + "description": "Legacy model retained for compatibility with older integrations", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-01-15", - "last_updated": "2025-01-15", + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2026-05-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 1000000 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.2, - "output": 1.1 + "input": 0.4, + "output": 2, + "cache_read": 0.04, + "cache_write": 0.5 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "MiniMax M2.1", - "family": "minimax", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { - "field": "reasoning_details" + "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.75, + "output": 3.5, + "cache_read": 0.15 } }, - "minimax/minimax-m2.5:free": { - "id": "minimax/minimax-m2.5:free", - "name": "MiniMax M2.5 (free)", - "family": "minimax", + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek-V3.2", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { - "field": "reasoning_details" + "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-11", + "release_date": "2026-01-20", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 163000, + "input": 160000, + "output": 16384 }, + "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 0.45, + "reasoning": 0.45, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "minimax/minimax-m1": { - "id": "minimax/minimax-m1", - "name": "MiniMax M1", - "family": "minimax", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2025-01", + "release_date": "2025-01-20", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 40000 + "context": 196608, + "input": 190000, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 2.2 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", + "MiniMaxAI/MiniMax-M2.5-fast": { + "id": "MiniMaxAI/MiniMax-M2.5-fast", + "name": "MiniMax-M2.5-fast", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-01", + "release_date": "2025-01-20", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 8000, + "input": 7000, + "output": 8192 }, + "status": "deprecated", "cost": { "input": 0.3, "output": 1.2, - "cache_read": 0.03 + "cache_read": 0.03, + "cache_write": 0.375 } }, - "qwen/qwen3-coder-plus": { - "id": "qwen/qwen3-coder-plus", - "name": "Qwen3 Coder Plus", - "family": "qwen", + "PrimeIntellect/INTELLECT-3": { + "id": "PrimeIntellect/INTELLECT-3", + "name": "INTELLECT-3", + "description": "Legacy model retained for compatibility with older integrations", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "knowledge": "2025-10", + "release_date": "2026-01-25", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 65536 + "context": 128000, + "input": 120000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.65, - "output": 3.25, - "cache_read": 0.13, - "cache_write": 0.8125 + "input": 0.2, + "output": 1.1, + "cache_read": 0.02, + "cache_write": 0.25 } - }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", - "family": "qwen", - "attachment": true, + } + } + }, + "auriko": { + "id": "auriko", + "env": ["AURIKO_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.auriko.ai/v1", + "name": "Auriko", + "doc": "https://docs.auriko.ai", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "qwen/qwen2.5-vl-72b-instruct": { - "id": "qwen/qwen2.5-vl-72b-instruct", - "name": "Qwen2.5 VL 72B Instruct", - "family": "qwen", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-02-01", - "last_updated": "2025-02-01", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "qwen/qwen-plus": { - "id": "qwen/qwen-plus", - "name": "Qwen: Qwen-Plus", - "family": "qwen", - "attachment": false, - "reasoning": false, + "grok-4.3": { + "id": "grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 32768 + "output": 30000 }, "cost": { - "input": 0.26, - "output": 0.78, - "cache_read": 0.052, - "cache_write": 0.325 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "qwen/qwen3.5-flash-02-23": { - "id": "qwen/qwen3.5-flash-02-23", - "name": "Qwen: Qwen3.5-Flash", - "family": "qwen", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-25", - "last_updated": "2026-02-25", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 1048576, "output": 65536 }, "cost": { - "input": 0.065, - "output": 0.26 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03 } }, - "qwen/qwen3.6-plus": { - "id": "qwen/qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen", - "attachment": true, + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 65536 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.325, - "output": 1.95, - "cache_read": 0.0325, - "cache_write": 0.40625 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "qwen/qwen3-max": { - "id": "qwen/qwen3-max", - "name": "Qwen3 Max", - "family": "qwen", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 + } + }, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 1.2, - "output": 6, - "cache_read": 0.156, - "cache_write": 0.975 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen/qwen3-coder:exacto": { - "id": "qwen/qwen3-coder:exacto", - "name": "Qwen3 Coder (exacto)", - "family": "qwen", + "minimax-m2-7-highspeed": { + "id": "minimax-m2-7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.38, - "output": 1.53 + "input": 0.6, + "output": 2.4, + "cache_write": 0.375 } }, - "qwen/qwen3-30b-a3b-instruct-2507": { - "id": "qwen/qwen3-30b-a3b-instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507", - "family": "qwen", + "minimax-m2-7": { + "id": "minimax-m2-7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.3, + "output": 1.2, + "cache_write": 0.375 } }, - "qwen/qwen-3.6-27b": { - "id": "qwen/qwen-3.6-27b", - "name": "Qwen3.6 27B", + "qwen-3.6-plus": { + "id": "qwen-3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", "family": "qwen", - "attachment": true, - "reasoning": false, + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 81920 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.195, - "output": 1.56 + "input": 0.5, + "output": 3, + "cache_read": 0.1, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } } }, - "qwen/qwen3-235b-a22b-thinking-2507": { - "id": "qwen/qwen3-235b-a22b-thinking-2507", - "name": "Qwen3 235B A22B Thinking 2507", - "family": "qwen", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 81920 + "output": 262144 }, "cost": { - "input": 0.078, - "output": 0.312 + "input": 0.5, + "output": 2.8 } }, - "qwen/qwen3-next-80b-a3b-thinking": { - "id": "qwen/qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", - "family": "qwen", - "attachment": false, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-11", - "last_updated": "2025-09-11", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, @@ -25321,469 +29139,747 @@ "output": 262144 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "qwen/qwen3-30b-a3b-thinking-2507": { - "id": "qwen/qwen3-30b-a3b-thinking-2507", - "name": "Qwen3 30B A3B Thinking 2507", - "family": "qwen", - "attachment": false, + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "qwen/qwen3-coder-flash": { - "id": "qwen/qwen3-coder-flash", - "name": "Qwen3 Coder Flash", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 66536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.5, - "cache_read": 0.039, - "cache_write": 0.24375 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen/qwen3-next-80b-a3b-instruct": { - "id": "qwen/qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "family": "qwen", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + } + } + }, + "stepfun-ai": { + "id": "stepfun-ai", + "env": ["STEPFUN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.stepfun.ai/step_plan/v1", + "name": "StepFun AI", + "doc": "https://platform.stepfun.ai/docs/en/step-plan/integrations/open-code", + "models": { + "step-2-16k": { + "id": "step-2-16k", + "name": "Step 2 (16K)", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-11", - "last_updated": "2025-09-11", + "knowledge": "2024-06", + "release_date": "2025-01-01", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 5.21, + "output": 16.44, + "cache_read": 1.04 } }, - "qwen/qwen-2.5-coder-32b-instruct": { - "id": "qwen/qwen-2.5-coder-32b-instruct", - "name": "Qwen2.5 Coder 32B Instruct", - "family": "qwen", + "step-tts-2": { + "id": "step-tts-2", + "name": "Step TTS 2", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "step", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, + "temperature": false, + "release_date": "2026-03-01", + "last_updated": "2026-07-02", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + }, + "step-3.5-flash": { + "id": "step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-11-11", - "last_updated": "2024-11-11", + "knowledge": "2025-01", + "release_date": "2026-01-29", + "last_updated": "2026-06-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0, + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 + } + }, + "stepaudio-2.5-asr": { + "id": "stepaudio-2.5-asr", + "name": "StepAudio 2.5 ASR", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "step", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-24", + "last_updated": "2026-07-02", + "modalities": { + "input": ["audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 0, "output": 0 } }, - "qwen/qwen3-coder-30b-a3b-instruct": { - "id": "qwen/qwen3-coder-30b-a3b-instruct", - "name": "Qwen3 Coder 30B A3B Instruct", - "family": "qwen", + "stepaudio-2.5-tts": { + "id": "stepaudio-2.5-tts", + "name": "StepAudio 2.5 TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "step", "attachment": false, "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-07-02", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + }, + "step-3.5-flash-2603": { + "id": "step-3.5-flash-2603", + "name": "Step 3.5 Flash 2603", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 160000, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.07, - "output": 0.27 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "qwen/qwen3-coder": { - "id": "qwen/qwen3-coder", - "name": "Qwen3 Coder", - "family": "qwen", - "attachment": false, - "reasoning": false, + "step-3.7-flash": { + "id": "step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-06-29", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 66536 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "qwen/qwen3.5-plus-02-15": { - "id": "qwen/qwen3.5-plus-02-15", - "name": "Qwen3.5 Plus 2026-02-15", - "family": "qwen", - "attachment": true, + "step-1-32k": { + "id": "step-1-32k", + "name": "Step 1 (32K)", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2024-06", + "release_date": "2025-01-01", + "last_updated": "2026-02-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.4, - "output": 2.4 + "input": 2.05, + "output": 9.59, + "cache_read": 0.41 } - }, - "qwen/qwen3-235b-a22b-07-25": { - "id": "qwen/qwen3-235b-a22b-07-25", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", + } + } + }, + "vivgrid": { + "id": "vivgrid", + "env": ["VIVGRID_API_KEY"], + "npm": "@ai-sdk/openai", + "api": "https://api.vivgrid.com/v1", + "name": "Vivgrid", + "doc": "https://docs.vivgrid.com/models", + "models": { + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-28", - "last_updated": "2025-07-21", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 131072 + "context": 1000000, + "output": 384000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.15, - "output": 0.85 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "google/gemini-2.5-pro-preview-05-06": { - "id": "google/gemini-2.5-pro-preview-05-06", - "name": "Gemini 2.5 Pro Preview 05-06", - "family": "gemini-pro", + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-05-06", - "last_updated": "2025-05-06", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "google/gemini-3.1-pro-preview-customtools": { - "id": "google/gemini-3.1-pro-preview-customtools", - "name": "Gemini 3.1 Pro Preview Custom Tools", - "family": "gemini-pro", - "attachment": true, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, "interleaved": { - "field": "reasoning_details" + "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 1.2, + "output": 4.2, + "cache_read": 0.3 + } + }, + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 2, - "output": 12, - "reasoning": 12, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "google/gemma-3-4b-it:free": { - "id": "google/gemma-3-4b-it:free", - "name": "Gemma 3 4B (free)", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "google/gemini-2.5-flash-lite-preview-09-2025": { - "id": "google/gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview 09-25", - "family": "gemini-flash-lite", - "attachment": true, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "google/gemini-2.0-flash-001": { - "id": "google/gemini-2.0-flash-001", - "name": "Gemini 2.0 Flash", - "family": "gemini-flash", + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT 5.6 Luna", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "google/gemma-3n-e4b-it": { - "id": "google/gemma-3n-e4b-it", - "name": "Gemma 3n 4B", - "family": "gemma", + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT 5.6 Terra", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.02, - "output": 0.04 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "google/gemini-3.1-flash-lite-preview": { - "id": "google/gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text", "image", "video", "pdf", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.25, - "output": 1.5, - "reasoning": 1.5, - "cache_read": 0.025, - "cache_write": 0.083, - "input_audio": 0.5, - "output_audio": 0.5 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "google/gemma-3n-e4b-it:free": { - "id": "google/gemma-3n-e4b-it:free", - "name": "Gemma 3n 4B (free)", - "family": "gemma", + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "google/gemini-3.1-pro-preview": { - "id": "google/gemini-3.1-pro-preview", + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, "structured_output": true, "temperature": true, "knowledge": "2025-01", "release_date": "2026-02-19", "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, @@ -25791,10 +29887,13 @@ "context": 1048576, "output": 65536 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { "input": 2, "output": 12, - "reasoning": 12, + "cache_read": 0.2, "tiers": [ { "input": 4, @@ -25813,1857 +29912,1823 @@ } } }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 + "input": 0.25, + "output": 2, + "cache_read": 0.03 } }, - "google/gemini-3-pro-preview": { - "id": "google/gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT 5.6 Sol", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1050000, - "output": 66000 - }, - "cost": { - "input": 2, - "output": 12 - } - }, - "google/gemma-3n-e2b-it:free": { - "id": "google/gemma-3n-e2b-it:free", - "name": "Gemma 3n 2B (free)", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 8192, - "output": 2000 + "input": 922000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "google/gemma-2-9b-it": { - "id": "google/gemma-2-9b-it", - "name": "Gemma 2 9B", - "family": "gemma", + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-28", - "last_updated": "2024-06-28", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 128000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.03, - "output": 0.09 + "input": 0.28, + "output": 0.42 } }, - "google/gemma-4-31b-it": { - "id": "google/gemma-4-31b-it", - "name": "Gemma 4 31B", - "family": "gemma", + "gemini-3.1-flash-lite-preview": { + "id": "gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.14, - "output": 0.4 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "cache_write": 1 } }, - "google/gemini-2.5-pro-preview-06-05": { - "id": "google/gemini-2.5-pro-preview-06-05", - "name": "Gemini 2.5 Pro Preview 06-05", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "google/gemma-3-12b-it": { - "id": "google/gemma-3-12b-it", - "name": "Gemma 3 12B", - "family": "gemma", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + } + } + }, + "tinfoil": { + "id": "tinfoil", + "env": ["TINFOIL_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://inference.tinfoil.sh/v1", + "name": "Tinfoil", + "doc": "https://docs.tinfoil.sh", + "models": { + "kimi-k2-6": { + "id": "kimi-k2-6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 256000, + "output": 262144 }, "cost": { - "input": 0.03, - "output": 0.1 + "input": 1.5, + "output": 5.25 } }, - "google/gemma-3-27b-it:free": { - "id": "google/gemma-3-27b-it:free", - "name": "Gemma 3 27B (free)", - "family": "gemma", + "llama3-3-70b": { + "id": "llama3-3-70b", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 1.75, + "output": 2.75 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, + "gpt-oss-safeguard-120b": { + "id": "gpt-oss-safeguard-120b", + "name": "gpt-oss-safeguard-120b", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-07-17", - "last_updated": "2025-07-17", + "knowledge": "2024-06", + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 131000, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.0375 + "input": 0.15, + "output": 0.6 } }, - "google/gemini-3.1-flash-image-preview": { - "id": "google/gemini-3.1-flash-image-preview", - "name": "Gemini 3.1 Flash Image Preview (Nano Banana 2)", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, + "nomic-embed-text": { + "id": "nomic-embed-text", + "name": "Nomic Embed Text v1.5", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "structured_output": false, + "temperature": false, + "release_date": "2024-02", + "last_updated": "2024-02", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 8192, + "output": 768 }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.05, + "output": 0 } }, - "google/gemma-4-31b-it:free": { - "id": "google/gemma-4-31b-it:free", - "name": "Gemma 4 31B (free)", - "family": "gemma", - "attachment": true, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "knowledge": "2024-06", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 131000, "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.6 } }, - "google/gemma-3-12b-it:free": { - "id": "google/gemma-3-12b-it:free", - "name": "Gemma 3 12B (free)", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, + "glm-5-2": { + "id": "glm-5-2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 8192 + "context": 384000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25 } }, - "google/gemma-3-4b-it": { - "id": "google/gemma-3-4b-it", - "name": "Gemma 3 4B", + "gemma4-31b": { + "id": "gemma4-31b", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", "family": "gemma", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 96000, - "output": 96000 + "context": 256000, + "output": 32768 }, "cost": { - "input": 0.01703, - "output": 0.06815 + "input": 0.4, + "output": 1 } - }, - "google/gemini-2.5-flash-preview-09-2025": { - "id": "google/gemini-2.5-flash-preview-09-2025", - "name": "Gemini 2.5 Flash Preview 09-25", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, + } + } + }, + "mistral": { + "id": "mistral", + "env": ["MISTRAL_API_KEY"], + "npm": "@ai-sdk/mistral", + "name": "Mistral", + "doc": "https://docs.mistral.ai/getting-started/models/", + "models": { + "codestral-latest": { + "id": "codestral-latest", + "name": "Codestral (latest)", + "description": "Mistral code model for completions, refactors, and developer IDE workflows", + "family": "codestral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "knowledge": "2024-10", + "release_date": "2024-05-29", + "last_updated": "2025-01-04", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 256000, + "output": 4096 }, "cost": { "input": 0.3, - "output": 2.5, - "cache_read": 0.031 + "output": 0.9 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Gemma 3 27B", - "family": "gemma", + "mistral-large-latest": { + "id": "mistral-large-latest", + "name": "Mistral Large (latest)", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 96000, - "output": 96000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.04, - "output": 0.15 + "input": 0.5, + "output": 1.5 } }, - "google/gemma-4-26b-a4b-it": { - "id": "google/gemma-4-26b-a4b-it", - "name": "Gemma 4 26B A4B", - "family": "gemma", - "attachment": true, - "reasoning": true, + "open-mistral-7b": { + "id": "open-mistral-7b", + "name": "Mistral 7B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-03", - "last_updated": "2026-04-03", + "knowledge": "2023-12", + "release_date": "2023-09-27", + "last_updated": "2023-09-27", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 8000, + "output": 8000 }, "cost": { - "input": 0.13, - "output": 0.4 + "input": 0.25, + "output": 0.25 } }, - "google/gemma-4-26b-a4b-it:free": { - "id": "google/gemma-4-26b-a4b-it:free", - "name": "Gemma 4 26B A4B (free)", - "family": "gemma", - "attachment": true, - "reasoning": true, + "devstral-small-2507": { + "id": "devstral-small-2507", + "name": "Devstral Small", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-03", - "last_updated": "2026-04-03", + "knowledge": "2025-05", + "release_date": "2025-07-10", + "last_updated": "2025-07-10", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 128000, + "output": 128000 }, + "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.3 } }, - "google/gemini-2.5-flash-lite": { - "id": "google/gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, + "ministral-3b-latest": { + "id": "ministral-3b-latest", + "name": "Ministral 3B (latest)", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2024-10", + "release_date": "2024-10-01", + "last_updated": "2024-10-04", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.04, + "output": 0.04 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "pixtral-large-latest": { + "id": "pixtral-large-latest", + "name": "Pixtral Large (latest)", + "description": "Mistral's larger vision model for document-heavy image understanding and chat", + "family": "pixtral", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2024-11-04", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 2, + "output": 6 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 Instruct 0905", - "family": "kimi", + "mistral-nemo": { + "id": "mistral-nemo", + "name": "Mistral Nemo", + "description": "Efficient Mistral-NVIDIA open model for multilingual chat and local deployment", + "family": "mistral-nemo", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.15, + "output": 0.15 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "mistral-embed": { + "id": "mistral-embed", + "name": "Mistral Embed", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "mistral-embed", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2023-12-11", + "last_updated": "2023-12-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 8000, + "output": 3072 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.1, + "output": 0 } }, - "moonshotai/kimi-k2-0905:exacto": { - "id": "moonshotai/kimi-k2-0905:exacto", - "name": "Kimi K2 Instruct 0905 (exacto)", - "family": "kimi", + "mistral-small-2506": { + "id": "mistral-small-2506", + "name": "Mistral Small 3.2", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2025-03", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.1, + "output": 0.3 } }, - "moonshotai/kimi-k2": { - "id": "moonshotai/kimi-k2", - "name": "Kimi K2", - "family": "kimi", + "ministral-8b-latest": { + "id": "ministral-8b-latest", + "name": "Ministral 8B (latest)", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2024-10", - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "release_date": "2024-10-01", + "last_updated": "2024-10-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 0.1, + "output": 0.1 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "open-mixtral-8x22b": { + "id": "open-mixtral-8x22b", + "name": "Mixtral 8x22B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mixtral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "knowledge": "2024-04", + "release_date": "2024-04-17", + "last_updated": "2024-04-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 64000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 2, + "output": 6 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "mistral-medium-latest": { + "id": "mistral-medium-latest", + "name": "Mistral Medium (latest)", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1.5, + "output": 7.5 } }, - "anthropic/claude-3.7-sonnet": { - "id": "anthropic/claude-3.7-sonnet", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "devstral-small-2505": { + "id": "devstral-small-2505", + "name": "Devstral Small 2505", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-01", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "knowledge": "2025-05", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, + "context": 128000, "output": 128000 }, + "status": "deprecated", "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.1, + "output": 0.3 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "magistral-small": { + "id": "magistral-small", + "name": "Magistral Small", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-small", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "knowledge": "2025-06", + "release_date": "2025-03-17", + "last_updated": "2025-03-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, + "context": 128000, "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 0.5, + "output": 1.5 } }, - "anthropic/claude-opus-4.7": { - "id": "anthropic/claude-opus-4.7", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "mistral-medium-2604": { + "id": "mistral-medium-2604", + "name": "Mistral Medium 3.5", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 1.5, + "output": 7.5 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "mistral-small-latest": { + "id": "mistral-small-latest", + "name": "Mistral Small (latest)", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.15, + "output": 0.6 } }, - "anthropic/claude-sonnet-4.5": { - "id": "anthropic/claude-sonnet-4.5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "open-mixtral-8x7b": { + "id": "open-mixtral-8x7b", + "name": "Mixtral 8x7B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mixtral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2024-01", + "release_date": "2023-12-11", + "last_updated": "2023-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 32000, + "output": 32000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.7, + "output": 0.7 } }, - "anthropic/claude-opus-4.5": { - "id": "anthropic/claude-opus-4.5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "devstral-latest": { + "id": "devstral-latest", + "name": "Devstral 2", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05-30", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.4, + "output": 2 } }, - "anthropic/claude-haiku-4.5": { - "id": "anthropic/claude-haiku-4.5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "mistral-small-2603": { + "id": "mistral-small-2603", + "name": "Mistral Small 4", + "description": "Fast Mistral production model for chat, extraction, and cost-sensitive agents", + "family": "mistral-small", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.15, + "output": 0.6 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "mistral-medium-2505": { + "id": "mistral-medium-2505", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "knowledge": "2025-05", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.4, + "output": 2 } }, - "deepseek/deepseek-v4-flash": { - "id": "deepseek/deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "mistral-large-2411": { + "id": "mistral-large-2411", + "name": "Mistral Large 2.1", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2024-11", + "release_date": "2024-11-18", + "last_updated": "2024-11-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 2, + "output": 6 } }, - "deepseek/deepseek-v4-pro": { - "id": "deepseek/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": true, + "mistral-medium-2508": { + "id": "mistral-medium-2508", + "name": "Mistral Medium 3.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-08-12", + "last_updated": "2025-08-12", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.4, + "output": 2 + } + }, + "open-mistral-nemo": { + "id": "open-mistral-nemo", + "name": "Open Mistral Nemo", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mistral-nemo", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 128000, + "output": 128000 }, + "status": "deprecated", "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0.15, + "output": 0.15 } }, - "x-ai/grok-4.3": { - "id": "x-ai/grok-4.3", - "name": "Grok 4.3", - "family": "grok", - "attachment": true, + "magistral-medium-latest": { + "id": "magistral-medium-latest", + "name": "Magistral Medium (latest)", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-medium", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-05-01", - "last_updated": "2026-05-01", + "knowledge": "2025-06", + "release_date": "2025-03-17", + "last_updated": "2025-03-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 1000000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2, - "context_over_200k": { - "input": 2.5, - "output": 5, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2.5, - "output": 5, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 2, + "output": 5 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, - "reasoning": true, + "devstral-medium-latest": { + "id": "devstral-medium-latest", + "name": "Devstral 2 (latest)", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.4, + "output": 2 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, - "reasoning": true, + "devstral-2512": { + "id": "devstral-2512", + "name": "Devstral 2", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.4, + "output": 2 } }, - "openai/gpt-5.5-pro": { - "id": "openai/gpt-5.5-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "labs-devstral-small-2512": { + "id": "labs-devstral-small-2512", + "name": "Devstral Small 2", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 256000, + "output": 256000 }, + "status": "deprecated", "cost": { - "input": 30, - "output": 180, - "context_over_200k": { - "input": 60, - "output": 270 - }, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0, + "output": 0 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + "pixtral-12b": { + "id": "pixtral-12b", + "name": "Pixtral 12B", + "description": "Mistral vision-language model for image understanding and multimodal chat", + "family": "pixtral", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "knowledge": "2024-09", + "release_date": "2024-09-01", + "last_updated": "2024-09-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - }, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.15, + "output": 0.15 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude Opus 4", - "family": "claude-opus", + "mistral-large-2512": { + "id": "mistral-large-2512", + "name": "Mistral Large 3", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.5, + "output": 1.5 } }, - "anthropic/claude-3.5-haiku": { - "id": "anthropic/claude-3.5-haiku", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", - "attachment": true, + "devstral-medium-2507": { + "id": "devstral-medium-2507", + "name": "Devstral Medium", + "description": "Legacy model retained for compatibility with older integrations", + "family": "devstral", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "knowledge": "2025-05", + "release_date": "2025-07-10", + "last_updated": "2025-07-10", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 128000, + "output": 128000 }, + "status": "deprecated", "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0.4, + "output": 2 } - }, - "xiaomi/mimo-v2.5-pro": { - "id": "xiaomi/mimo-v2.5-pro", - "name": "Xiaomi: MiMo-V2.5-Pro", - "family": "mimo", + } + } + }, + "cloudflare-workers-ai": { + "id": "cloudflare-workers-ai", + "env": ["CLOUDFLARE_ACCOUNT_ID", "CLOUDFLARE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.cloudflare.com/client/v4/accounts/${CLOUDFLARE_ACCOUNT_ID}/ai/v1", + "name": "Cloudflare Workers AI", + "doc": "https://developers.cloudflare.com/workers-ai/models/", + "models": { + "@cf/ibm-granite/granite-4.0-h-micro": { + "id": "@cf/ibm-granite/granite-4.0-h-micro", + "name": "Granite 4.0 H Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "granite", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2025-10-07", + "last_updated": "2025-10-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 131000, + "output": 131000 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.017, + "output": 0.112 } }, - "xiaomi/mimo-v2-omni": { - "id": "xiaomi/mimo-v2-omni", - "name": "Xiaomi: MiMo-V2-Omni", - "family": "mimo", + "@cf/moonshotai/kimi-k2.7-code": { + "id": "@cf/moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 262144, - "output": 131072 + "output": 262144 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "xiaomi/mimo-v2.5": { - "id": "xiaomi/mimo-v2.5", - "name": "Xiaomi: MiMo-V2.5", - "family": "mimo", + "@cf/moonshotai/kimi-k2.6": { + "id": "@cf/moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { - "field": "reasoning_details" + "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 262144, + "output": 256000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - }, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "xiaomi/mimo-v2-pro": { - "id": "xiaomi/mimo-v2-pro", - "name": "Xiaomi: MiMo-V2-Pro", - "family": "mimo", - "attachment": false, + "@cf/google/gemma-4-26b-a4b-it": { + "id": "@cf/google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, + "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 256000, + "output": 16384 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.1, + "output": 0.3 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "Xiaomi: MiMo-V2-Flash", - "family": "mimo", + "@cf/openai/gpt-oss-120b": { + "id": "@cf/openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 0.35, + "output": 0.75 } - } - } - }, - "fireworks-ai": { - "id": "fireworks-ai", - "env": ["FIREWORKS_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.fireworks.ai/inference/v1/", - "name": "Fireworks AI", - "doc": "https://fireworks.ai/docs/", - "models": { - "accounts/fireworks/models/glm-5p1": { - "id": "accounts/fireworks/models/glm-5p1", - "name": "GLM 5.1", - "family": "glm", + }, + "@cf/openai/gpt-oss-20b": { + "id": "@cf/openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202800, - "output": 131072 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.2, + "output": 0.3 } }, - "accounts/fireworks/models/deepseek-v3p2": { - "id": "accounts/fireworks/models/deepseek-v3p2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "@cf/mistralai/mistral-small-3.1-24b-instruct": { + "id": "@cf/mistralai/mistral-small-3.1-24b-instruct", + "name": "Mistral Small 3.1 24B Instruct", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2025-03-18", + "last_updated": "2025-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 160000, - "output": 160000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.56, - "output": 1.68, - "cache_read": 0.28 + "input": 0.351, + "output": 0.555 } }, - "accounts/fireworks/models/minimax-m2p5": { - "id": "accounts/fireworks/models/minimax-m2p5", - "name": "MiniMax-M2.5", - "family": "minimax", + "@cf/nvidia/nemotron-3-120b-a12b": { + "id": "@cf/nvidia/nemotron-3-120b-a12b", + "name": "Nemotron 3 Super 120B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "interleaved": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 196608 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 0.5, + "output": 1.5 } }, - "accounts/fireworks/models/glm-4p5-air": { - "id": "accounts/fireworks/models/glm-4p5-air", - "name": "GLM 4.5 Air", - "family": "glm-air", + "@cf/zai-org/glm-5.2": { + "id": "@cf/zai-org/glm-5.2", + "name": "Glm 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.22, - "output": 0.88 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "accounts/fireworks/models/glm-5": { - "id": "accounts/fireworks/models/glm-5", - "name": "GLM 5", - "family": "glm", + "@cf/zai-org/glm-4.7-flash": { + "id": "@cf/zai-org/glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, + "context": 131072, "output": 131072 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.5 + "input": 0.0605, + "output": 0.4 } }, - "accounts/fireworks/models/deepseek-v3p1": { - "id": "accounts/fireworks/models/deepseek-v3p1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b": { + "id": "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b", + "name": "Deepseek R1 Distill Qwen 32B", + "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 163840 + "context": 80000, + "output": 80000 }, "cost": { - "input": 0.56, - "output": 1.68 + "input": 0.497, + "output": 4.881 } }, - "accounts/fireworks/models/kimi-k2p6": { - "id": "accounts/fireworks/models/kimi-k2p6", - "name": "Kimi K2.6", - "family": "kimi-thinking", + "@cf/qwen/qwen3-30b-a3b-fp8": { + "id": "@cf/qwen/qwen3-30b-a3b-fp8", + "name": "Qwen3 30B A3b fp8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "release_date": "2025-04-30", + "last_updated": "2025-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.0509, + "output": 0.335 } }, - "accounts/fireworks/models/kimi-k2-instruct": { - "id": "accounts/fireworks/models/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", + "@cf/qwen/qwen2.5-coder-32b-instruct": { + "id": "@cf/qwen/qwen2.5-coder-32b-instruct", + "name": "Qwen2.5 Coder 32B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1, - "output": 3 + "input": 0.66, + "output": 1 } }, - "accounts/fireworks/models/qwen3p6-plus": { - "id": "accounts/fireworks/models/qwen3p6-plus", - "name": "Qwen 3.6 Plus", + "@cf/qwen/qwq-32b": { + "id": "@cf/qwen/qwq-32b", + "name": "Qwq 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "family": "qwen", - "attachment": true, + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2026-04-04", - "last_updated": "2026-04-04", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 24000, + "output": 24000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.1 + "input": 0.66, + "output": 1 } }, - "accounts/fireworks/models/minimax-m2p1": { - "id": "accounts/fireworks/models/minimax-m2p1", - "name": "MiniMax-M2.1", - "family": "minimax", + "@cf/meta/llama-3.2-1b-instruct": { + "id": "@cf/meta/llama-3.2-1b-instruct", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 60000, + "output": 60000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 0.027, + "output": 0.201 } }, - "accounts/fireworks/models/minimax-m2p7": { - "id": "accounts/fireworks/models/minimax-m2p7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "@cf/meta/llama-3.2-11b-vision-instruct": { + "id": "@cf/meta/llama-3.2-11b-vision-instruct", + "name": "Llama 3.2 11B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2026-04-12", - "last_updated": "2026-04-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 196608 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 0.0485, + "output": 0.676 } }, - "accounts/fireworks/models/glm-4p7": { - "id": "accounts/fireworks/models/glm-4p7", - "name": "GLM 4.7", - "family": "glm", - "attachment": false, - "reasoning": true, + "@cf/meta/llama-4-scout-17b-16e-instruct": { + "id": "@cf/meta/llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E Instruct", + "description": "Open Llama with long-context vision for efficient multimodal agents", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, - "output": 198000 + "context": 131000, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.3 + "input": 0.27, + "output": 0.85 } }, - "accounts/fireworks/models/glm-4p5": { - "id": "accounts/fireworks/models/glm-4p5", - "name": "GLM 4.5", - "family": "glm", + "@cf/meta/llama-guard-3-8b": { + "id": "@cf/meta/llama-guard-3-8b", + "name": "Llama Guard 3 8B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "release_date": "2025-01-22", + "last_updated": "2025-01-22", "modalities": { "input": ["text"], "output": ["text"] @@ -27674,312 +31739,35825 @@ "output": 131072 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0.484, + "output": 0.03 } }, - "accounts/fireworks/models/kimi-k2p5": { - "id": "accounts/fireworks/models/kimi-k2p5", - "name": "Kimi K2.5", - "family": "kimi-thinking", + "@cf/meta/llama-3.3-70b-instruct-fp8-fast": { + "id": "@cf/meta/llama-3.3-70b-instruct-fp8-fast", + "name": "Llama 3.3 70B Instruct fp8 Fast", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 24000, + "output": 24000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.293, + "output": 2.253 } }, - "accounts/fireworks/models/gpt-oss-20b": { - "id": "accounts/fireworks/models/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + "@cf/meta/llama-3.1-8b-instruct-fp8": { + "id": "@cf/meta/llama-3.1-8b-instruct-fp8", + "name": "Llama 3.1 8B Instruct fp8", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-07-25", + "last_updated": "2024-07-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.05, - "output": 0.2 + "input": 0.152, + "output": 0.287 } }, - "accounts/fireworks/models/gpt-oss-120b": { - "id": "accounts/fireworks/models/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "@cf/meta/llama-3.2-3b-instruct": { + "id": "@cf/meta/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 80000, + "output": 80000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.0509, + "output": 0.335 } }, - "accounts/fireworks/models/kimi-k2-thinking": { - "id": "accounts/fireworks/models/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "@cf/aisingapore/gemma-sea-lion-v4-27b-it": { + "id": "@cf/aisingapore/gemma-sea-lion-v4-27b-it", + "name": "Gemma Sea Lion V4 27B It", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.3 + "input": 0.351, + "output": 0.555 } - }, - "accounts/fireworks/routers/kimi-k2p5-turbo": { - "id": "accounts/fireworks/routers/kimi-k2p5-turbo", - "name": "Kimi K2.5 Turbo", - "family": "kimi-thinking", + } + } + }, + "bailing": { + "id": "bailing", + "env": ["BAILING_API_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.tbox.cn/api/llm/v1/chat/completions", + "name": "Bailing", + "doc": "https://alipaytbox.yuque.com/sxs0ba/ling/intro", + "models": { + "Ring-1T": { + "id": "Ring-1T", + "name": "Ring-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "ring", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2024-06", + "release_date": "2025-10", + "last_updated": "2025-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.57, + "output": 2.29 } }, - "accounts/fireworks/models/deepseek-v4-pro": { - "id": "accounts/fireworks/models/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "Ling-1T": { + "id": "Ling-1T", + "name": "Ling-1T", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "ling", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2024-06", + "release_date": "2025-10", + "last_updated": "2025-10", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 128000, + "output": 32000 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.15 + "input": 0.57, + "output": 2.29 } } } }, - "kimi-for-coding": { - "id": "kimi-for-coding", - "env": ["KIMI_API_KEY"], - "npm": "@ai-sdk/anthropic", - "api": "https://api.kimi.com/coding/v1", - "name": "Kimi For Coding", - "doc": "https://www.kimi.com/coding/docs/en/third-party-agents.html", + "anyapi": { + "id": "anyapi", + "env": ["ANYAPI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.anyapi.ai/v1", + "name": "AnyAPI", + "doc": "https://docs.anyapi.ai", "models": { - "k2p6": { - "id": "k2p6", - "name": "Kimi K2.6", - "family": "kimi-thinking", - "attachment": false, + "xai/grok-4.3": { + "id": "xai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04", - "last_updated": "2026-04", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 - }, - "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "context": 1000000, + "output": 30000 } }, - "k2p5": { - "id": "k2p5", - "name": "Kimi K2.5", - "family": "kimi-thinking", - "attachment": false, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 - }, - "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "context": 1048576, + "output": 65536 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11", - "last_updated": "2025-12", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 1048576, + "output": 65536 + } + }, + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + } + }, + "google/gemini-3-pro-preview": { + "id": "google/gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + } + }, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + } + }, + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + } + }, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + } + }, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + } + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + } + }, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + } + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + } + }, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "mistralai/devstral-2512": { + "id": "mistralai/devstral-2512", + "name": "Devstral 2", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "status": "deprecated" + }, + "mistralai/mistral-large-2512": { + "id": "mistralai/mistral-large-2512", + "name": "Mistral Large 3", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + } + }, + "anthropic/claude-sonnet-4-5": { + "id": "anthropic/claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + } + }, + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + } + }, + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + } + }, + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + } + }, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + } + }, + "cohere/command-r-plus-08-2024": { + "id": "cohere/command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4000 + } + }, + "perplexity/sonar-reasoning-pro": { + "id": "perplexity/sonar-reasoning-pro", + "name": "Sonar Reasoning Pro", + "description": "Web-grounded Sonar for multi-step research questions that need cited reasoning", + "family": "sonar-reasoning", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + } + }, + "perplexity/sonar-pro": { + "id": "perplexity/sonar-pro", + "name": "Sonar Pro", + "description": "Deeper Sonar search model with broader retrieval and stronger synthesis", + "family": "sonar-pro", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + } + }, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + } + }, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + } + }, + "deepseek/deepseek-r1": { + "id": "deepseek/deepseek-r1", + "name": "DeepSeek Reasoner", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + } + }, + "deepseek/deepseek-chat": { + "id": "deepseek/deepseek-chat", + "name": "DeepSeek Chat", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + } + } + } + }, + "google": { + "id": "google", + "env": ["GOOGLE_API_KEY", "GOOGLE_GENERATIVE_AI_API_KEY", "GEMINI_API_KEY"], + "npm": "@ai-sdk/google", + "name": "Google", + "doc": "https://ai.google.dev/gemini-api/docs/models", + "models": { + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "gemini-2.5-flash-preview-tts": { + "id": "gemini-2.5-flash-preview-tts", + "name": "Gemini 2.5 Flash Preview TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gemini-flash", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-05-01", + "last_updated": "2025-05-01", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 16384 + }, + "cost": { + "input": 0.5, + "output": 10 + } + }, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + } + }, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 + } + }, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 + } + }, + "gemma-4-31b-it": { + "id": "gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + } + }, + "gemini-2.0-flash": { + "id": "gemini-2.0-flash", + "name": "Gemini 2.0 Flash", + "description": "Earlier Gemini Flash workhorse for responsive multimodal apps and tool use", + "family": "gemini-flash", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 8192 + }, + "status": "deprecated", + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "gemini-embedding-001": { + "id": "gemini-embedding-001", + "name": "Gemini Embedding 001", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "gemini", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2025-05", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2048, + "output": 1 + }, + "cost": { + "input": 0.15, + "output": 0 + } + }, + "gemini-3.1-pro-preview-customtools": { + "id": "gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "gemini-flash-lite-latest": { + "id": "gemini-flash-lite-latest", + "name": "Gemini Flash-Lite Latest", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "gemini-3-pro-image-preview": { + "id": "gemini-3-pro-image-preview", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 120 + } + }, + "gemini-2.5-flash-image": { + "id": "gemini-2.5-flash-image", + "name": "Nano Banana", + "description": "Nano Banana image model for fast generation, edits, and character-consistent assets", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 32768 + }, + "cost": { + "input": 0.3, + "output": 30, + "cache_read": 0.075 + } + }, + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 + } + }, + "gemini-omni-flash-preview": { + "id": "gemini-omni-flash-preview", + "name": "Gemini Omni Flash Preview", + "description": "Video generation and editing model for fast, conversational text- and image-to-video workflows", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image", "video"], + "output": ["video"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 17.5 + } + }, + "gemini-3.1-flash-image-preview": { + "id": "gemini-3.1-flash-image-preview", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 65536, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 60 + } + }, + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "gemma-4-26b-a4b-it": { + "id": "gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + } + }, + "gemini-3-pro-preview": { + "id": "gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "status": "deprecated", + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 + } + }, + "gemini-2.5-pro-preview-tts": { + "id": "gemini-2.5-pro-preview-tts", + "name": "Gemini 2.5 Pro Preview TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gemini-flash", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-05-01", + "last_updated": "2025-05-01", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 16384 + }, + "cost": { + "input": 1, + "output": 20 + } + }, + "gemini-flash-latest": { + "id": "gemini-flash-latest", + "name": "Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.075, + "input_audio": 1 + } + }, + "gemini-3.1-flash-lite-preview": { + "id": "gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "status": "deprecated", + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "gemini-2.0-flash-lite": { + "id": "gemini-2.0-flash-lite", + "name": "Gemini 2.0 Flash-Lite", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 8192 + }, + "status": "deprecated", + "cost": { + "input": 0.075, + "output": 0.3 + } + } + } + }, + "opencode-go": { + "id": "opencode-go", + "env": ["OPENCODE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://opencode.ai/zen/go/v1", + "name": "OpenCode Go", + "doc": "https://opencode.ai/docs/zen", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 + } + }, + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax-m2.5", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 65536 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 + } + }, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.7-plus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.04, + "cache_write": 0.5, + "tiers": [ + { + "input": 1.2, + "output": 4.8, + "cache_read": 0.12, + "cache_write": 1.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.2, + "output": 4.8, + "cache_read": 0.12, + "cache_write": 1.5 + } + } + }, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "qwen3.7-max", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 + } + }, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.19 + } + }, + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 32768 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 + } + }, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 1.74, + "output": 3.48, + "cache_read": 0.0145 + } + }, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 + } + }, + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax-m3", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-31", + "last_updated": "2026-05-31", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "tiers": [ + { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 + } + } + }, + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen3.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 65536 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.2, + "output": 1.2, + "cache_read": 0.02, + "cache_write": 0.25 + } + }, + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax-m2.7", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 + } + }, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "status": "deprecated", + "cost": { + "input": 0.6, + "output": 3, + "cache_read": 0.1 + } + }, + "mimo-v2.5": { + "id": "mimo-v2.5", + "name": "MiMo V2.5", + "description": "MiMo omni model for text, image, video, audio, and agents", + "family": "mimo-v2.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 + } + }, + "mimo-v2-omni": { + "id": "mimo-v2-omni", + "name": "MiMo V2 Omni", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo-v2-omni", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text", "image", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 128000 + }, + "status": "deprecated", + "cost": { + "input": 0.4, + "output": 2, + "cache_read": 0.08 + } + }, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.16 + } + }, + "mimo-v2-pro": { + "id": "mimo-v2-pro", + "name": "MiMo V2 Pro", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo-v2-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 128000 + }, + "status": "deprecated", + "cost": { + "input": 1, + "output": 3, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } + } + }, + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo V2.5 Pro", + "description": "MiMo pro model for strong multimodal reasoning and agent execution", + "family": "mimo-v2.5-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 128000 + }, + "cost": { + "input": 1.74, + "output": 3.48, + "cache_read": 0.0145 + } + }, + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 32768 + }, + "status": "deprecated", + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.2 + } + }, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/anthropic" + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } + } + } + } + }, + "digitalocean": { + "id": "digitalocean", + "env": ["DIGITALOCEAN_ACCESS_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://inference.do-ai.run/v1", + "name": "DigitalOcean", + "doc": "https://docs.digitalocean.com/products/gradient-ai-platform/details/models/", + "models": { + "anthropic-claude-haiku-4.5": { + "id": "anthropic-claude-haiku-4.5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 1, + "cache_write": 1.25 + } + }, + "openai-gpt-image-1": { + "id": "openai-gpt-image-1", + "name": "GPT Image 1", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-04-24", + "last_updated": "2025-04-24", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + }, + "cost": { + "input": 5, + "output": 40, + "cache_read": 1.25 + } + }, + "e5-large-v2": { + "id": "e5-large-v2", + "name": "E5 Large v2", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2023-05-19", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512, + "output": 1024 + }, + "cost": { + "input": 0.02, + "output": 0 + } + }, + "bge-m3": { + "id": "bge-m3", + "name": "BGE M3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "bge", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-01-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 1024 + }, + "cost": { + "input": 0.02, + "output": 0 + } + }, + "mistral-3-14B": { + "id": "mistral-3-14B", + "name": "Ministral 3 14B Instruct", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-15", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 0.2 + } + }, + "nemotron-3-ultra-550b": { + "id": "nemotron-3-ultra-550b", + "name": "Nemotron 3 Ultra", + "description": "Flagship Nemotron model for high-throughput reasoning and complex agents", + "family": "nemotron", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-06-04", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + } + }, + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax-m2.5", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08", + "release_date": "2026-02-12", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 128000 + }, + "status": "beta", + "cost": { + "input": 0.3, + "output": 1.2 + } + }, + "openai-gpt-5.4-nano": { + "id": "openai-gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 + } + }, + "deepseek-v3": { + "id": "deepseek-v3", + "name": "DeepSeek V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-12-26", + "last_updated": "2025-03-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 163840, + "output": 131072 + } + }, + "openai-gpt-image-2": { + "id": "openai-gpt-image-2", + "name": "GPT Image 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-04-24", + "last_updated": "2025-04-24", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + }, + "openai-gpt-5.2": { + "id": "openai-gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "deepseek-r1-distill-llama-70b": { + "id": "deepseek-r1-distill-llama-70b", + "name": "DeepSeek R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-01-30", + "last_updated": "2025-01-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.99, + "output": 0.99 + } + }, + "qwen3-embedding-0.6b": { + "id": "qwen3-embedding-0.6b", + "name": "Qwen3 Embedding 0.6B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-06-03", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8000, + "output": 1024 + }, + "status": "beta", + "cost": { + "input": 0.04, + "output": 0 + } + }, + "gemma-4-31B-it": { + "id": "gemma-4-31B-it", + "name": "Gemma 4 31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.18, + "output": 0.5 + } + }, + "llama-4-maverick": { + "id": "llama-4-maverick", + "name": "Llama 4 Maverick 17B 128E Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 16384 + }, + "cost": { + "input": 0.25, + "output": 0.87 + } + }, + "anthropic-claude-3.7-sonnet": { + "id": "anthropic-claude-3.7-sonnet", + "name": "Claude 3.7 Sonnet", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2025-02-24", + "last_updated": "2025-02-24", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "status": "deprecated", + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "openai-gpt-4o-mini": { + "id": "openai-gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "anthropic-claude-opus-4.7": { + "id": "anthropic-claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "deepseek-4-flash": { + "id": "deepseek-4-flash", + "name": "Deepseek V4 Flash", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-05-27", + "last_updated": "2026-05-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 8192 + } + }, + "anthropic-claude-4.5-haiku": { + "id": "anthropic-claude-4.5-haiku", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 1, + "cache_write": 1.25 + } + }, + "anthropic-claude-4.6-sonnet": { + "id": "anthropic-claude-4.6-sonnet", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75 + } + } + }, + "anthropic-claude-sonnet-4": { + "id": "anthropic-claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75 + } + } + }, + "ministral-3-8b-instruct-2512": { + "id": "ministral-3-8b-instruct-2512", + "name": "Ministral 3 8B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-12-15", + "last_updated": "2025-12-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + } + }, + "anthropic-claude-4.5-sonnet": { + "id": "anthropic-claude-4.5-sonnet", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.3, + "cache_write": 3.75 + } + } + }, + "anthropic-claude-opus-4.6": { + "id": "anthropic-claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 0.5, + "cache_write": 6.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 0.5, + "cache_write": 6.25 + } + } + }, + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0.975, + "output": 4.3, + "cache_read": 0.26 + } + }, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 393216 + }, + "cost": { + "input": 1.74, + "output": 3.48 + } + }, + "openai-gpt-oss-20b": { + "id": "openai-gpt-oss-20b", + "name": "gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-08-05", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.05, + "output": 0.45 + } + }, + "qwen-2.5-14b-instruct": { + "id": "qwen-2.5-14b-instruct", + "name": "Qwen 2.5 14B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + } + }, + "anthropic-claude-opus-4.8": { + "id": "anthropic-claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-28", + "last_updated": "2026-05-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.05, + "output": 4.4, + "cache_read": 0.21 + } + }, + "anthropic-claude-opus-4.5": { + "id": "anthropic-claude-opus-4.5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic-claude-4.1-opus": { + "id": "anthropic-claude-4.1-opus", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "llama3-8b-instruct": { + "id": "llama3-8b-instruct", + "name": "Llama 3.1 Instruct (8B)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.198, + "output": 0.198 + } + }, + "stable-diffusion-3.5-large": { + "id": "stable-diffusion-3.5-large", + "name": "Stable Diffusion 3.5 Large", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "stable-diffusion", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-10-22", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["image"] + }, + "open_weights": true, + "limit": { + "context": 256, + "output": 1 + }, + "cost": { + "input": 0.08, + "output": 0 + } + }, + "openai-gpt-5.4-pro": { + "id": "openai-gpt-5.4-pro", + "name": "GPT-5.4 pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180 + } + }, + "all-mini-lm-l6-v2": { + "id": "all-mini-lm-l6-v2", + "name": "All-MiniLM-L6-v2", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2021-08-30", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256, + "output": 384 + }, + "cost": { + "input": 0.009, + "output": 0 + } + }, + "openai-gpt-image-1.5": { + "id": "openai-gpt-image-1.5", + "name": "GPT Image 1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-11-25", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + }, + "cost": { + "input": 5, + "output": 10, + "cache_read": 1 + } + }, + "openai-gpt-4o": { + "id": "openai-gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "anthropic-claude-opus-4": { + "id": "anthropic-claude-opus-4", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "bge-reranker-v2-m3": { + "id": "bge-reranker-v2-m3", + "name": "BGE Reranker v2 M3", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "bge", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-03-12", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 1 + }, + "cost": { + "input": 0.01, + "output": 0 + } + }, + "openai-gpt-5": { + "id": "openai-gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai-o3-mini": { + "id": "openai-o3-mini", + "name": "o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "multi-qa-mpnet-base-dot-v1": { + "id": "multi-qa-mpnet-base-dot-v1", + "name": "Multi-QA-mpnet-base-dot-v1", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2021-08-30", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512, + "output": 768 + }, + "cost": { + "input": 0.009, + "output": 0 + } + }, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.5, + "output": 2.7 + } + }, + "llama3.3-70b-instruct": { + "id": "llama3.3-70b-instruct", + "name": "Llama 3.3 Instruct 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 128000 + }, + "cost": { + "input": 0.65, + "output": 0.65 + } + }, + "gte-large-en-v1.5": { + "id": "gte-large-en-v1.5", + "name": "GTE Large (v1.5)", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-03-27", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 1024 + }, + "cost": { + "input": 0.09, + "output": 0 + } + }, + "openai-gpt-5.4-mini": { + "id": "openai-gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "openai-gpt-5.5": { + "id": "openai-gpt-5.5", + "name": "GPT-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + }, + "nemotron-3-nano-30b": { + "id": "nemotron-3-nano-30b", + "name": "Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + } + }, + "openai-gpt-oss-120b": { + "id": "openai-gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-08-05", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.1, + "output": 0.7 + } + }, + "openai-gpt-5-nano": { + "id": "openai-gpt-5-nano", + "name": "GPT-5 nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 + } + }, + "wan2-2-t2v-a14b": { + "id": "wan2-2-t2v-a14b", + "name": "Wan2.2-T2V-A14B", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-07-28", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["video"] + }, + "open_weights": true, + "limit": { + "context": 100, + "output": 1 + }, + "cost": { + "input": 0.6, + "output": 0 + } + }, + "mistral-7b-instruct-v0.3": { + "id": "mistral-7b-instruct-v0.3", + "name": "Mistral 7B Instruct v0.3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2024-05-22", + "last_updated": "2024-05-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 32768 + } + }, + "qwen3-tts-voicedesign": { + "id": "qwen3-tts-voicedesign", + "name": "Qwen3 TTS VoiceDesign", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 1 + } + }, + "mistral-nemo-instruct-2407": { + "id": "mistral-nemo-instruct-2407", + "name": "Mistral Nemo Instruct", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mistral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "status": "deprecated", + "cost": { + "input": 0.3, + "output": 0.3 + } + }, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4 + } + }, + "deepseek-3.2": { + "id": "deepseek-3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-12-02", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.5, + "output": 1.6 + } + }, + "openai-o1": { + "id": "openai-o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 15, + "output": 60, + "cache_read": 7.5 + } + }, + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen 3.5 397B A17B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen3.5", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-15", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 81920 + }, + "cost": { + "input": 0.55, + "output": 3.5 + } + }, + "qwen3-coder-flash": { + "id": "qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.45, + "output": 1.7 + } + }, + "openai-gpt-5.3-codex": { + "id": "openai-gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai-gpt-4.1": { + "id": "openai-gpt-4.1", + "name": "GPT-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "anthropic-claude-3.5-haiku": { + "id": "anthropic-claude-3.5-haiku", + "name": "Claude 3.5 Haiku", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-haiku", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-11-05", + "last_updated": "2024-11-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "status": "deprecated", + "cost": { + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 + } + }, + "openai-gpt-5.4": { + "id": "openai-gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + }, + "arcee-trinity-large-thinking": { + "id": "arcee-trinity-large-thinking", + "name": "Trinity Large Thinking", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "trinity", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 128000 + }, + "status": "beta", + "cost": { + "input": 0.25, + "output": 0.9, + "cache_read": 0.06 + } + }, + "openai-o3": { + "id": "openai-o3", + "name": "o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "anthropic-claude-3.5-sonnet": { + "id": "anthropic-claude-3.5-sonnet", + "name": "Claude 3.5 Sonnet", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-sonnet", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-06-20", + "last_updated": "2024-10-22", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "status": "deprecated", + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "openai-gpt-5-mini": { + "id": "openai-gpt-5-mini", + "name": "GPT-5 mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "nvidia-nemotron-3-super-120b": { + "id": "nvidia-nemotron-3-super-120b", + "name": "Nemotron-3-Super-120B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2026-02", + "release_date": "2026-03-11", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 32768 + }, + "status": "beta", + "cost": { + "input": 0.3, + "output": 0.65 + } + }, + "glm-5": { + "id": "glm-5", + "name": "GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "release_date": "2026-02-11", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 128000 + }, + "cost": { + "input": 1, + "output": 3.2 + } + }, + "openai-gpt-5.2-pro": { + "id": "openai-gpt-5.2-pro", + "name": "GPT-5.2 pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 21, + "output": 168 + } + }, + "anthropic-claude-3-opus": { + "id": "anthropic-claude-3-opus", + "name": "Claude 3 Opus", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-opus", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-08", + "release_date": "2024-02-29", + "last_updated": "2024-02-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 4096 + }, + "status": "deprecated", + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic-claude-fable-5": { + "id": "anthropic-claude-fable-5", + "name": "Anthropic Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-09", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "nemotron-3-nano-omni": { + "id": "nemotron-3-nano-omni", + "name": "Nemotron Nano 3 Omni", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 65536, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 0.9 + } + }, + "alibaba-qwen3-32b": { + "id": "alibaba-qwen3-32b", + "name": "Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131000, + "output": 40960 + }, + "cost": { + "input": 0.25, + "output": 0.55 + } + }, + "nemotron-nano-12b-v2-vl": { + "id": "nemotron-nano-12b-v2-vl", + "name": "Nemotron Nano 12B v2 VL", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "family": "nemotron", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-12-01", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.2, + "output": 0.6 + } + }, + "openai-gpt-5.1-codex-max": { + "id": "openai-gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "fal-ai/fast-sdxl": { + "id": "fal-ai/fast-sdxl", + "name": "Fast SDXL", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "stable-diffusion", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2023-07-26", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["image"] + }, + "open_weights": true, + "limit": { + "context": 0, + "output": 0 + } + }, + "fal-ai/elevenlabs/tts/multilingual-v2": { + "id": "fal-ai/elevenlabs/tts/multilingual-v2", + "name": "ElevenLabs Multilingual TTS v2", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "elevenlabs", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2023-08-22", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + }, + "fal-ai/flux/schnell": { + "id": "fal-ai/flux/schnell", + "name": "FLUX.1 [schnell]", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-08-01", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["image"] + }, + "open_weights": true, + "limit": { + "context": 0, + "output": 0 + } + }, + "fal-ai/stable-audio-25/text-to-audio": { + "id": "fal-ai/stable-audio-25/text-to-audio", + "name": "Stable Audio 2.5 (Text-to-Audio)", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-10-08", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + } + } + }, + "subconscious": { + "id": "subconscious", + "env": ["SUBCONSCIOUS_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.subconscious.dev/v1", + "name": "Subconscious", + "doc": "https://docs.subconscious.dev", + "models": { + "subconscious/glm-5.2": { + "id": "subconscious/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 + } + }, + "subconscious/tim-qwen3.6-27b": { + "id": "subconscious/tim-qwen3.6-27b", + "name": "TIM-Qwen3.6 27B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "input": 8192, + "output": 5000 + }, + "cost": { + "input": 0.3, + "output": 3, + "cache_read": 0.15 + } + } + } + }, + "venice": { + "id": "venice", + "env": ["VENICE_API_KEY"], + "npm": "venice-ai-sdk-provider", + "name": "Venice AI", + "doc": "https://docs.venice.ai", + "models": { + "z-ai-glm-5-turbo": { + "id": "z-ai-glm-5-turbo", + "name": "GLM 5 Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 32768 + }, + "cost": { + "input": 1.2, + "output": 4, + "cache_read": 0.24 + } + }, + "grok-4-20-multi-agent": { + "id": "grok-4-20-multi-agent", + "name": "Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "release_date": "2026-03-12", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 128000 + }, + "cost": { + "input": 1.42, + "output": 2.83, + "cache_read": 0.23, + "tiers": [ + { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45 + } + } + }, + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 32768 + }, + "cost": { + "input": 0.138, + "output": 0.275, + "cache_read": 0.028 + } + }, + "google-gemma-4-31b-it": { + "id": "google-gemma-4-31b-it", + "name": "Google Gemma 4 31B Instruct", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-03", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.12, + "output": 0.36, + "cache_read": 0.09 + } + }, + "kimi-k2-6": { + "id": "kimi-k2-6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-20", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.75, + "output": 3.5, + "cache_read": 0.16 + } + }, + "openai-gpt-56-terra-pro": { + "id": "openai-gpt-56-terra-pro", + "name": "GPT-5.6 Terra Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 3.125, + "output": 18.75, + "cache_read": 0.3125, + "cache_write": 3.90625 + } + }, + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "Qwen 3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2025-04-29", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.75 + } + }, + "openai-gpt-56-sol-pro": { + "id": "openai-gpt-56-sol-pro", + "name": "GPT-5.6 Sol Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 6.25, + "output": 37.5, + "cache_read": 0.625, + "cache_write": 7.8125 + } + }, + "nvidia-nemotron-cascade-2-30b-a3b": { + "id": "nvidia-nemotron-cascade-2-30b-a3b", + "name": "Nemotron Cascade 2 30B A3B", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 32768 + }, + "cost": { + "input": 0.14, + "output": 0.8 + } + }, + "claude-opus-4-7-fast": { + "id": "claude-opus-4-7-fast", + "name": "Claude Opus 4.7 Fast", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-05-14", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 36, + "output": 180, + "cache_read": 3.6, + "cache_write": 45 + } + }, + "openai-gpt-55-pro": { + "id": "openai-gpt-55-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 37.5, + "output": 225 + } + }, + "llama-3.3-70b": { + "id": "llama-3.3-70b", + "name": "Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "release_date": "2025-04-06", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0.7, + "output": 2.8 + } + }, + "qwen3-5-397b-a17b": { + "id": "qwen3-5-397b-a17b", + "name": "Qwen 3.5 397B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-16", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0.75, + "output": 4.5 + } + }, + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-12-06", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 198000, + "output": 32768 + }, + "cost": { + "input": 6, + "output": 30, + "cache_read": 0.6, + "cache_write": 7.5 + } + }, + "zai-org-glm-5": { + "id": "zai-org-glm-5", + "name": "GLM 5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-11", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 32000 + }, + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.2 + } + }, + "qwen3-5-35b-a3b": { + "id": "qwen3-5-35b-a3b", + "name": "Qwen 3.5 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-25", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 16384 + }, + "cost": { + "input": 0.3125, + "output": 1.25, + "cache_read": 0.15625 + } + }, + "venice-uncensored-role-play": { + "id": "venice-uncensored-role-play", + "name": "Venice Role Play Uncensored", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "venice", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-20", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 2 + } + }, + "qwen-3-6-plus": { + "id": "qwen-3-6-plus", + "name": "Qwen 3.6 Plus Uncensored", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-06", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.625, + "output": 3.75, + "cache_read": 0.0625, + "cache_write": 0.78, + "tiers": [ + { + "input": 2.5, + "output": 7.5, + "cache_read": 0.0625, + "cache_write": 0.78, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 7.5, + "cache_read": 0.0625, + "cache_write": 0.78 + } + } + }, + "zai-org-glm-4.6": { + "id": "zai-org-glm-4.6", + "name": "GLM 4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2024-04-01", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 16384 + }, + "cost": { + "input": 0.43, + "output": 1.75, + "cache_read": 0.08 + } + }, + "openai-gpt-4o-2024-11-20": { + "id": "openai-gpt-4o-2024-11-20", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2026-02-28", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 3.125, + "output": 12.5 + } + }, + "grok-4-3": { + "id": "grok-4-3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-18", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 32000 + }, + "cost": { + "input": 1.42, + "output": 2.83, + "cache_read": 0.23, + "tiers": [ + { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45 + } + } + }, + "qwen-3-7-plus": { + "id": "qwen-3-7-plus", + "name": "Qwen 3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 2, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 1.5, + "output": 6, + "cache_read": 0.15, + "cache_write": 1.875, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.5, + "output": 6, + "cache_read": 0.15, + "cache_write": 1.875 + } + } + }, + "minimax-m27": { + "id": "minimax-m27", + "name": "MiniMax M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 32768 + }, + "cost": { + "input": 0.375, + "output": 1.5, + "cache_read": 0.06875 + } + }, + "kimi-k2-7-code": { + "id": "kimi-k2-7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-13", + "last_updated": "2026-06-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.75, + "output": 3.5, + "cache_read": 0.16 + } + }, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 32768 + }, + "cost": { + "input": 1.65, + "output": 3.301, + "cache_read": 0.33 + } + }, + "qwen3-235b-a22b-thinking-2507": { + "id": "qwen3-235b-a22b-thinking-2507", + "name": "Qwen 3 235B A22B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "release_date": "2025-04-29", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.45, + "output": 3.5 + } + }, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-01-15", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 198000, + "output": 64000 + }, + "cost": { + "input": 3.75, + "output": 18.75, + "cache_read": 0.375, + "cache_write": 4.69 + } + }, + "openai-gpt-4o-mini-2024-07-18": { + "id": "openai-gpt-4o-mini-2024-07-18", + "name": "GPT-4o Mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2026-02-28", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.1875, + "output": 0.75, + "cache_read": 0.09375 + } + }, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 6, + "output": 30, + "cache_read": 0.6, + "cache_write": 7.5 + } + }, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-29", + "last_updated": "2026-07-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "openai-gpt-53-codex": { + "id": "openai-gpt-53-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 2.19, + "output": 17.5, + "cache_read": 0.219 + } + }, + "openai-gpt-54": { + "id": "openai-gpt-54", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 131072 + }, + "cost": { + "input": 3.13, + "output": 18.8, + "cache_read": 0.313 + } + }, + "openai-gpt-56-terra": { + "id": "openai-gpt-56-terra", + "name": "GPT-5.6 Terra", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 3.125, + "output": 18.75, + "cache_read": 0.3125, + "cache_write": 3.90625 + } + }, + "google-gemma-3-27b-it": { + "id": "google-gemma-3-27b-it", + "name": "Google Gemma 3 27B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2025-11-04", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 16384 + }, + "cost": { + "input": 0.12, + "output": 0.2 + } + }, + "openai-gpt-56-luna": { + "id": "openai-gpt-56-luna", + "name": "GPT-5.6 Luna", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 7.5, + "cache_read": 0.125, + "cache_write": 1.5625 + } + }, + "aion-labs-aion-3-0-mini": { + "id": "aion-labs-aion-3-0-mini", + "name": "Aion 3.0 Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-07-08", + "last_updated": "2026-07-08", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0.875, + "output": 1.75, + "cache_read": 0.225 + } + }, + "claude-opus-4-8-fast": { + "id": "claude-opus-4-8-fast", + "name": "Claude Opus 4.8 Fast", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 12, + "output": 60, + "cache_read": 1.2, + "cache_write": 15 + } + }, + "gemma-4-uncensored": { + "id": "gemma-4-uncensored", + "name": "Gemma 4 Uncensored", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-04-13", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.1625, + "output": 0.5 + } + }, + "qwen3-next-80b": { + "id": "qwen3-next-80b", + "name": "Qwen 3 Next 80b", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2025-04-29", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 16384 + }, + "cost": { + "input": 0.35, + "output": 1.9 + } + }, + "mistral-small-2603": { + "id": "mistral-small-2603", + "name": "Mistral Small 4", + "description": "Fast Mistral production model for chat, extraction, and cost-sensitive agents", + "family": "mistral-small", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.1875, + "output": 0.75 + } + }, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 6, + "output": 30, + "cache_read": 0.6, + "cache_write": 7.5 + } + }, + "zai-org-glm-4.7-flash": { + "id": "zai-org-glm-4.7-flash", + "name": "GLM 4.7 Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-29", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.125, + "output": 0.5 + } + }, + "google-gemma-4-26b-a4b-it": { + "id": "google-gemma-4-26b-a4b-it", + "name": "Google Gemma 4 26B A4B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.13, + "output": 0.4, + "cache_read": 0.05 + } + }, + "grok-4-5": { + "id": "grok-4-5", + "name": "Grok 4.5", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-07-07", + "last_updated": "2026-07-08", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 500000, + "output": 32000 + }, + "cost": { + "input": 2.27, + "output": 6.8, + "cache_read": 0.57, + "tiers": [ + { + "input": 4.53, + "output": 13.6, + "cache_read": 1.13, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4.53, + "output": 13.6, + "cache_read": 1.13 + } + } + }, + "grok-4-20": { + "id": "grok-4-20", + "name": "Grok 4.20", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-03-12", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 128000 + }, + "cost": { + "input": 1.42, + "output": 2.83, + "cache_read": 0.23, + "tiers": [ + { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.83, + "output": 5.67, + "cache_read": 0.45 + } + } + }, + "hermes-3-llama-3.1-405b": { + "id": "hermes-3-llama-3.1-405b", + "name": "Hermes 3 Llama 3.1 405b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "hermes", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2025-09-25", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.1, + "output": 3 + } + }, + "aion-labs-aion-2-0": { + "id": "aion-labs-aion-2-0", + "name": "Aion 2.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "release_date": "2026-03-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 1, + "output": 2, + "cache_read": 0.25 + } + }, + "qwen3-vl-235b-a22b": { + "id": "qwen3-vl-235b-a22b", + "name": "Qwen3 VL 235B", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-01-16", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.21, + "output": 1.9, + "cache_read": 0.1 + } + }, + "xiaomi-mimo-v2-5": { + "id": "xiaomi-mimo-v2-5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-06-11", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.05 + } + }, + "venice-uncensored-1-2": { + "id": "venice-uncensored-1-2", + "name": "Venice Uncensored 1.2", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "venice", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-04-01", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.2, + "output": 0.9 + } + }, + "gemini-3-5-flash": { + "id": "gemini-3-5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-22", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 1.55, + "output": 9.45, + "cache_read": 0.155, + "cache_write": 0.086 + } + }, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-10", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 12, + "output": 60, + "cache_read": 1.2, + "cache_write": 15 + } + }, + "openai-gpt-oss-120b": { + "id": "openai-gpt-oss-120b", + "name": "OpenAI GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-11-06", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.07, + "output": 0.3 + } + }, + "zai-org-glm-5-2": { + "id": "zai-org-glm-5-2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-16", + "last_updated": "2026-06-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 + } + }, + "nvidia-nemotron-3-nano-30b-a3b": { + "id": "nvidia-nemotron-3-nano-30b-a3b", + "name": "NVIDIA Nemotron 3 Nano 30B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.075, + "output": 0.3 + } + }, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 6, + "output": 30, + "cache_read": 0.6, + "cache_write": 7.5 + } + }, + "mistral-small-3-2-24b-instruct": { + "id": "mistral-small-3-2-24b-instruct", + "name": "Mistral Small 3.2 24B Instruct", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-01-15", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 16384 + }, + "cost": { + "input": 0.09375, + "output": 0.25 + } + }, + "openai-gpt-54-pro": { + "id": "openai-gpt-54-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 37.5, + "output": 225, + "tiers": [ + { + "input": 75, + "output": 337.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 75, + "output": 337.5 + } + } + }, + "minimax-m3-preview": { + "id": "minimax-m3-preview", + "name": "MiniMax M3 Preview", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax-m3", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "release_date": "2026-06-12", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 524288, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 + } + }, + "gemini-3-1-pro-preview": { + "id": "gemini-3-1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 32768 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.5, + "cache_write": 0.5, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 0.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 0.5 + } + } + }, + "qwen3-coder-480b-a35b-instruct-turbo": { + "id": "qwen3-coder-480b-a35b-instruct-turbo", + "name": "Qwen 3 Coder 480B Turbo", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "release_date": "2026-01-27", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.35, + "output": 1.5, + "cache_read": 0.04 + } + }, + "qwen3-6-27b": { + "id": "qwen3-6-27b", + "name": "Qwen 3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.325, + "output": 3.25 + } + }, + "openai-gpt-52-codex": { + "id": "openai-gpt-52-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08", + "release_date": "2025-01-15", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "input": 272000, + "output": 65536 + }, + "cost": { + "input": 2.19, + "output": 17.5, + "cache_read": 0.219 + } + }, + "qwen-3-7-max": { + "id": "qwen-3-7-max", + "name": "Qwen 3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-22", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 2.7, + "output": 8.05, + "cache_read": 0.27, + "cache_write": 3.35 + } + }, + "olafangensan-glm-4.7-flash-heretic": { + "id": "olafangensan-glm-4.7-flash-heretic", + "name": "GLM 4.7 Flash Heretic", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-04", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 24000 + }, + "cost": { + "input": 0.07, + "output": 0.4, + "cache_read": 0.035 + } + }, + "openai-gpt-54-mini": { + "id": "openai-gpt-54-mini", + "name": "GPT-5.4 Mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-27", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.9375, + "output": 5.625, + "cache_read": 0.09375 + } + }, + "grok-build-0-1": { + "id": "grok-build-0-1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 1, + "output": 2, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 4, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 4, + "cache_read": 0.4 + } + } + }, + "zai-org-glm-4.7": { + "id": "zai-org-glm-4.7", + "name": "GLM 4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-24", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 16384 + }, + "cost": { + "input": 0.55, + "output": 2.65, + "cache_read": 0.11 + } + }, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3.6, + "output": 18, + "cache_read": 0.36, + "cache_write": 4.5 + } + }, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-19", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.7, + "output": 3.75, + "cache_read": 0.07 + } + }, + "openai-gpt-55": { + "id": "openai-gpt-55", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 131072 + }, + "cost": { + "input": 6.25, + "output": 37.5, + "cache_read": 0.625, + "tiers": [ + { + "input": 12.5, + "output": 56.25, + "cache_read": 1.25, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 12.5, + "output": 56.25, + "cache_read": 1.25 + } + } + }, + "qwen3-5-9b": { + "id": "qwen3-5-9b", + "name": "Qwen 3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-05", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.15 + } + }, + "mercury-2": { + "id": "mercury-2", + "name": "Mercury 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "mercury", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-20", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 50000 + }, + "cost": { + "input": 0.3125, + "output": 0.9375, + "cache_read": 0.03125 + } + }, + "zai-org-glm-5-1": { + "id": "zai-org-glm-5-1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 80000 + }, + "cost": { + "input": 1.54, + "output": 4.84, + "cache_read": 0.286 + } + }, + "openai-gpt-56-luna-pro": { + "id": "openai-gpt-56-luna-pro", + "name": "GPT-5.6 Luna Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 7.5, + "cache_read": 0.125, + "cache_write": 1.5625 + } + }, + "kimi-k2-5": { + "id": "kimi-k2-5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2024-04", + "release_date": "2026-01-27", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 65536 + }, + "cost": { + "input": 0.56, + "output": 3.5, + "cache_read": 0.22 + } + }, + "z-ai-glm-5v-turbo": { + "id": "z-ai-glm-5v-turbo", + "name": "GLM 5V Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32768 + }, + "cost": { + "input": 1.5, + "output": 5, + "cache_read": 0.3 + } + }, + "nvidia-nemotron-3-ultra-550b-a55b": { + "id": "nvidia-nemotron-3-ultra-550b-a55b", + "name": "NVIDIA Nemotron 3 Ultra", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-04", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 32768 + }, + "cost": { + "input": 0.625, + "output": 3.125, + "cache_read": 0.1875 + } + }, + "aion-labs-aion-3-0": { + "id": "aion-labs-aion-3-0", + "name": "Aion 3.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-07-08", + "last_updated": "2026-07-08", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 3.75, + "output": 7.5, + "cache_read": 0.9375 + } + }, + "openai-gpt-52": { + "id": "openai-gpt-52", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-13", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "input": 272000, + "output": 65536 + }, + "cost": { + "input": 2.19, + "output": 17.5, + "cache_read": 0.219 + } + }, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2025-12-04", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 160000, + "output": 32768 + }, + "cost": { + "input": 0.33, + "output": 0.48, + "cache_read": 0.16 + } + }, + "openai-gpt-56-sol": { + "id": "openai-gpt-56-sol", + "name": "GPT-5.6 Sol", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 6.25, + "output": 37.5, + "cache_read": 0.625, + "cache_write": 7.8125 + } + }, + "minimax-m25": { + "id": "minimax-m25", + "name": "MiniMax M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 198000, + "output": 32768 + }, + "cost": { + "input": 0.27, + "output": 0.95, + "cache_read": 0.03 + } + }, + "llama-3.2-3b": { + "id": "llama-3.2-3b", + "name": "Llama 3.2 3B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "release_date": "2024-10-03", + "last_updated": "2026-06-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.6 + } + } + } + }, + "crossmodel": { + "id": "crossmodel", + "env": ["CROSSMODEL_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.crossmodel.ai/v1", + "name": "CrossModel", + "doc": "https://www.crossmodel.ai/docs", + "models": { + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.47, + "output": 2.16, + "cache_read": 0.1, + "cache_write": 0.47, + "tiers": [ + { + "input": 0.62, + "output": 2.47, + "cache_read": 0.13, + "cache_write": 0.62, + "tier": { + "type": "context", + "size": 32000 + } + } + ] + } + }, + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 1, + "output": 3.8, + "cache_read": 0.2, + "cache_write": 1, + "tiers": [ + { + "input": 1.2, + "output": 4.4, + "cache_read": 0.3, + "cache_write": 1.2, + "tier": { + "type": "context", + "size": 32000 + } + } + ] + } + }, + "z-ai/glm-5.2": { + "id": "z-ai/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 1.2, + "output": 4.4, + "cache_read": 0.3, + "cache_write": 1.2 + } + }, + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.6, + "output": 3, + "cache_read": 0.16, + "cache_write": 0.6, + "tiers": [ + { + "input": 0.8, + "output": 3.4, + "cache_read": 0.2, + "cache_write": 0.8, + "tier": { + "type": "context", + "size": 32000 + } + } + ] + } + }, + "z-ai/glm-5-turbo": { + "id": "z-ai/glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.9, + "output": 3.7, + "cache_read": 0.18, + "cache_write": 0.9, + "tiers": [ + { + "input": 1.1, + "output": 4.3, + "cache_read": 0.27, + "cache_write": 1.1, + "tier": { + "type": "context", + "size": 32000 + } + } + ] + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02, + "cache_write": 0.2 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 2.5, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 5 + } + } + }, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075, + "cache_write": 0.75 + } + }, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075, + "cache_write": 0.15 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 10, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 10 + } + } + }, + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0.16, + "output": 0.32, + "cache_read": 0.004, + "cache_write": 0.16 + } + }, + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0.47, + "output": 0.94, + "cache_read": 0.005, + "cache_write": 0.47 + } + }, + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-sonnet-5": { + "id": "anthropic/claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 + } + }, + "anthropic/claude-opus-4-8": { + "id": "anthropic/claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-fable-5": { + "id": "anthropic/claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 + } + }, + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "tencent/hy3-preview": { + "id": "tencent/hy3-preview", + "name": "Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 128000 + }, + "cost": { + "input": 0.19, + "output": 0.63, + "cache_read": 0.063, + "cache_write": 0.19, + "tiers": [ + { + "input": 0.25, + "output": 1, + "cache_read": 0.094, + "cache_write": 0.25, + "tier": { + "type": "context", + "size": 16000 + } + }, + { + "input": 0.32, + "output": 1.25, + "cache_read": 0.125, + "cache_write": 0.32, + "tier": { + "type": "context", + "size": 32000 + } + } + ] + } + }, + "moonshot/kimi-k2.7-code": { + "id": "moonshot/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 1, + "output": 4.16, + "cache_read": 0.18, + "cache_write": 1 + } + }, + "moonshot/kimi-k2.5": { + "id": "moonshot/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.62, + "output": 3.3, + "cache_read": 0.11, + "cache_write": 0.62 + } + }, + "moonshot/kimi-k2.6": { + "id": "moonshot/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 1, + "output": 4.16, + "cache_read": 0.18, + "cache_write": 1 + } + }, + "gemini/gemini-2.5-pro": { + "id": "gemini/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "cache_write": 1.25, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 2.5 + } + } + }, + "gemini/gemini-2.5-flash": { + "id": "gemini/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "cache_write": 0.3 + } + }, + "gemini/gemini-3.5-flash": { + "id": "gemini/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "cache_write": 1.5 + } + }, + "gemini/gemini-2.5-flash-lite": { + "id": "gemini/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0.1 + } + }, + "gemini/gemini-3.1-pro-preview": { + "id": "gemini/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "cache_write": 2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "cache_write": 4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4, + "cache_write": 4 + } + } + }, + "gemini/gemini-3-flash-preview": { + "id": "gemini/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.5 + } + }, + "qwen/qwen3.7-plus": { + "id": "qwen/qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0.32, + "output": 1.25, + "cache_read": 0.032, + "cache_write": 0.4, + "tiers": [ + { + "input": 0.96, + "output": 3.75, + "cache_read": 0.096, + "cache_write": 1.2, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.96, + "output": 3.75, + "cache_read": 0.096, + "cache_write": 1.2 + } + } + }, + "qwen/qwen3.7-max": { + "id": "qwen/qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 1.88, + "output": 5.63, + "cache_read": 0.375, + "cache_write": 2.35 + } + }, + "qwen/qwen3.6-flash": { + "id": "qwen/qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.19, + "output": 1.13, + "cache_read": 0.019, + "cache_write": 0.24, + "tiers": [ + { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075, + "cache_write": 0.94, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075, + "cache_write": 0.94 + } + } + }, + "qwen/qwen3.6-plus": { + "id": "qwen/qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.32, + "output": 1.88, + "cache_read": 0.032, + "cache_write": 0.4, + "tiers": [ + { + "input": 1.25, + "output": 7.5, + "cache_read": 0.124, + "cache_write": 1.57, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.25, + "output": 7.5, + "cache_read": 0.124, + "cache_write": 1.57 + } + } + }, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 65000 + }, + "cost": { + "input": 0.16, + "output": 0.32, + "cache_read": 0.004, + "cache_write": 0.16 + } + }, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 65000 + }, + "cost": { + "input": 0.47, + "output": 0.94, + "cache_read": 0.005, + "cache_write": 0.47 + } + }, + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1024000, + "output": 512000 + }, + "cost": { + "input": 0.33, + "output": 1.32, + "cache_read": 0.066, + "cache_write": 0.33, + "tiers": [ + { + "input": 0.66, + "output": 2.63, + "cache_read": 0.132, + "cache_write": 0.66, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.66, + "output": 2.63, + "cache_read": 0.132, + "cache_write": 0.66 + } + } + }, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.33, + "output": 1.32, + "cache_read": 0.066, + "cache_write": 0.42 + } + } + } + }, + "lmstudio": { + "id": "lmstudio", + "env": ["LMSTUDIO_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "http://127.0.0.1:1234/v1", + "name": "LMStudio", + "doc": "https://lmstudio.ai/models", + "models": { + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "qwen/qwen3-30b-a3b-2507": { + "id": "qwen/qwen3-30b-a3b-2507", + "name": "Qwen3 30B A3B 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "qwen/qwen3-coder-30b": { + "id": "qwen/qwen3-coder-30b", + "name": "Qwen3 Coder 30B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 + } + } + } + }, + "poolside": { + "id": "poolside", + "env": ["POOLSIDE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://inference.poolside.ai/v1", + "name": "Poolside", + "doc": "https://platform.poolside.ai", + "models": { + "poolside/laguna-xs.2": { + "id": "poolside/laguna-xs.2", + "name": "Laguna XS.2", + "description": "Agentic coding model from Poolside in the XS size class for local deployment", + "family": "laguna", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": false, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "poolside/laguna-m.1": { + "id": "poolside/laguna-m.1", + "name": "Laguna M.1", + "description": "Poolside's flagship agentic coding model for long-horizon work", + "family": "laguna", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": false, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "poolside/laguna-xs-2.1": { + "id": "poolside/laguna-xs-2.1", + "name": "Laguna XS 2.1", + "description": "Agentic coding model from Poolside in the XS size class for local deployment", + "family": "laguna", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": false, + "temperature": true, + "release_date": "2026-07-02", + "last_updated": "2026-07-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + } + } + }, + "zenifra": { + "id": "zenifra", + "env": ["ZENIFRA_AI_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://ai.zenifra.com/v1", + "name": "Zenifra", + "doc": "https://docs.zenifra.com", + "models": { + "alibaba/qwen3.6-35b-a3b": { + "id": "alibaba/qwen3.6-35b-a3b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "provider": { + "shape": "completions" + }, + "cost": { + "input": 0.19, + "output": 0.48 + } + } + } + }, + "zenmux": { + "id": "zenmux", + "env": ["ZENMUX_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://zenmux.ai/api/v1", + "name": "ZenMux", + "doc": "https://docs.zenmux.ai", + "models": { + "inclusionai/ling-1t": { + "id": "inclusionai/ling-1t", + "name": "Ling-1T", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-10-09", + "last_updated": "2025-10-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.56, + "output": 2.24, + "cache_read": 0.11 + } + }, + "inclusionai/ring-2.6-1t": { + "id": "inclusionai/ring-2.6-1t", + "name": "inclusionAI: Ring-2.6-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-12-31", + "release_date": "2026-05-07", + "last_updated": "2026-05-14", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 65000 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.06 + } + }, + "inclusionai/ring-1t": { + "id": "inclusionai/ring-1t", + "name": "Ring-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-10-12", + "last_updated": "2025-10-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.56, + "output": 2.24, + "cache_read": 0.11 + } + }, + "moonshotai/kimi-k2.7-code-free": { + "id": "moonshotai/kimi-k2.7-code-free", + "name": "Kimi K2.7 Code (Free)", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0 + } + }, + "moonshotai/kimi-k2-thinking-turbo": { + "id": "moonshotai/kimi-k2-thinking-turbo", + "name": "Kimi K2 Thinking Turbo", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 64000 + }, + "cost": { + "input": 1.15, + "output": 8, + "cache_read": 0.15 + } + }, + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.16 + } + }, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 64000 + }, + "cost": { + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 + } + }, + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2025-01-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 64000 + }, + "cost": { + "input": 0.58, + "output": 3.02, + "cache_read": 0.1 + } + }, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2025-01-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262140, + "output": 262140 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.16 + } + }, + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-04", + "last_updated": "2025-09-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 64000 + }, + "cost": { + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 + } + }, + "baidu/ernie-5.0-thinking-preview": { + "id": "baidu/ernie-5.0-thinking-preview", + "name": "ERNIE 5.0", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.84, + "output": 3.37 + } + }, + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 + } + }, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["pdf", "image", "text", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048000, + "output": 64000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.31, + "cache_write": 4.5 + } + }, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["pdf", "image", "text", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048000, + "output": 64000 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.07, + "cache_write": 1 + } + }, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + } + }, + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", + "modalities": { + "input": ["pdf", "image", "text", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048000, + "output": 64000 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.03, + "cache_write": 1 + } + }, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-02-19", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "pdf", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048000, + "output": 64000 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "cache_write": 4.5 + } + }, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "pdf", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048000, + "output": 64000 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 1 + } + }, + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-03-20", + "last_updated": "2025-03-20", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "output": 65530 + }, + "cost": { + "input": 0.25, + "output": 1.5 + } + }, + "x-ai/grok-code-fast-1": { + "id": "x-ai/grok-code-fast-1", + "name": "Grok Code Fast 1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.2, + "output": 1.5, + "cache_read": 0.02 + } + }, + "x-ai/grok-4.3": { + "id": "x-ai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 1000000 + }, + "cost": { + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "cache_write": 0, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "cache_write": 0, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "cache_write": 0 + } + } + }, + "x-ai/grok-4.1-fast-non-reasoning": { + "id": "x-ai/grok-4.1-fast-non-reasoning", + "name": "Grok 4.1 Fast Non Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 64000 + }, + "cost": { + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 + } + }, + "x-ai/grok-4": { + "id": "x-ai/grok-4", + "name": "Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.75 + } + }, + "x-ai/grok-4-fast": { + "id": "x-ai/grok-4-fast", + "name": "Grok 4 Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 64000 + }, + "cost": { + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 + } + }, + "x-ai/grok-4.2-fast-non-reasoning": { + "id": "x-ai/grok-4.2-fast-non-reasoning", + "name": "Grok 4.2 Fast Non Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 30000 + }, + "cost": { + "input": 3, + "output": 9 + } + }, + "x-ai/grok-4.2-fast": { + "id": "x-ai/grok-4.2-fast", + "name": "Grok 4.2 Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 30000 + }, + "cost": { + "input": 3, + "output": 9 + } + }, + "x-ai/grok-4.1-fast": { + "id": "x-ai/grok-4.1-fast", + "name": "Grok 4.1 Fast", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 64000 + }, + "cost": { + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 + } + }, + "x-ai/grok-build-0.1": { + "id": "x-ai/grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 1, + "output": 2, + "cache_read": 0.2 + } + }, + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.28, + "output": 1.14, + "cache_read": 0.06 + } + }, + "z-ai/glm-4.5": { + "id": "z-ai/glm-4.5", + "name": "GLM 4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.35, + "output": 1.54, + "cache_read": 0.07 + } + }, + "z-ai/glm-4.7-flashx": { + "id": "z-ai/glm-4.7-flashx", + "name": "GLM 4.7 FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.07, + "output": 0.42, + "cache_read": 0.01 + } + }, + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-03", + "last_updated": "2026-04-03", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0.8781, + "output": 3.5126, + "cache_read": 0.1903 + } + }, + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.35, + "output": 1.54, + "cache_read": 0.07 + } + }, + "z-ai/glm-5.2": { + "id": "z-ai/glm-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.5, + "cache_read": 0.26 + } + }, + "z-ai/glm-4.6v-flash-free": { + "id": "z-ai/glm-4.6v-flash-free", + "name": "GLM 4.6V Flash (Free)", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "z-ai/glm-4.7-flash-free": { + "id": "z-ai/glm-4.7-flash-free", + "name": "GLM 4.7 Flash (Free)", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "z-ai/glm-5.2-free": { + "id": "z-ai/glm-5.2-free", + "name": "GLM 5.2 (Free)", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0 + } + }, + "z-ai/glm-4.6v": { + "id": "z-ai/glm-4.6v", + "name": "GLM 4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.14, + "output": 0.42, + "cache_read": 0.03 + } + }, + "z-ai/glm-5v-turbo": { + "id": "z-ai/glm-5v-turbo", + "name": "GLM 5V Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", + "modalities": { + "input": ["text", "image", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.726, + "output": 3.1946, + "cache_read": 0.1743 + } + }, + "z-ai/glm-4.6v-flash": { + "id": "z-ai/glm-4.6v-flash", + "name": "GLM 4.6V FlashX", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.02, + "output": 0.21, + "cache_read": 0.0043 + } + }, + "z-ai/glm-4.5-air": { + "id": "z-ai/glm-4.5-air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.11, + "output": 0.56, + "cache_read": 0.02 + } + }, + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.58, + "output": 2.6, + "cache_read": 0.14 + } + }, + "z-ai/glm-5-turbo": { + "id": "z-ai/glm-5-turbo", + "name": "GLM 5 Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.88, + "output": 3.48 + } + }, + "openai/gpt-5.5-instant": { + "id": "openai/gpt-5.5-instant", + "name": "GPT-5.5 Instant", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-12-01", + "release_date": "2026-05-05", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 400000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + } + }, + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT-5.2-Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 21, + "output": 168 + } + }, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12 + } + }, + "openai/gpt-5.1-chat": { + "id": "openai/gpt-5.1-chat", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["pdf", "image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12 + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 0.2, + "output": 1.25 + } + }, + "openai/gpt-5.3-chat": { + "id": "openai/gpt-5.3-chat", + "name": "GPT-5.3 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16380 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.75, + "output": 14 + } + }, + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12 + } + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-01-01", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["image", "text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.17 + } + }, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.75, + "output": 14 + } + }, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1-Codex-Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.03 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 3.75, + "output": 18.75 + } + }, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 0.75, + "output": 4.5 + } + }, + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 45, + "output": 225 + } + }, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12 + } + }, + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-01-01", + "release_date": "2026-01-15", + "last_updated": "2026-01-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.17 + } + }, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["image", "text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "https://zenmux.ai/api/v1" + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + }, + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0.4, + "output": 2, + "cache_read": 0.08, + "tiers": [ + { + "input": 0.8, + "output": 4, + "cache_read": 0.16, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.8, + "output": 4, + "cache_read": 0.16 + } + } + }, + "xiaomi/mimo-v2-omni": { + "id": "xiaomi/mimo-v2-omni", + "name": "MiMo V2 Omni", + "description": "MiMo omni model for text, image, video, audio, and agents", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 265000, + "output": 265000 + }, + "cost": { + "input": 0.4, + "output": 2, + "cache_read": 0.08 + } + }, + "xiaomi/mimo-v2-flash": { + "id": "xiaomi/mimo-v2-flash", + "name": "MiMo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12-01", + "release_date": "2025-12-16", + "last_updated": "2026-02-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.3, + "cache_read": 0.01 + } + }, + "xiaomi/mimo-v2-pro": { + "id": "xiaomi/mimo-v2-pro", + "name": "MiMo V2 Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 256000 + }, + "cost": { + "input": 1, + "output": 3, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } + } + }, + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 1, + "output": 3, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } + } + }, + "sapiens-ai/agnes-1.5-pro": { + "id": "sapiens-ai/agnes-1.5-pro", + "name": "Agnes 1.5 Pro", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-21", + "last_updated": "2026-03-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 0.16, + "output": 0.8 + } + }, + "sapiens-ai/agnes-1.5-lite": { + "id": "sapiens-ai/agnes-1.5-lite", + "name": "Agnes 1.5 Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-03-26", + "last_updated": "2026-03-26", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 0.12, + "output": 0.6 + } + }, + "volcengine/doubao-seed-2.0-pro": { + "id": "volcengine/doubao-seed-2.0-pro", + "name": "Doubao-Seed-2.0-pro", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-02-14", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.45, + "output": 2.24, + "cache_read": 0.09, + "cache_write": 0.0024 + } + }, + "volcengine/doubao-seed-code": { + "id": "volcengine/doubao-seed-code", + "name": "Doubao-Seed-Code", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-11", + "last_updated": "2025-11-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.17, + "output": 1.12, + "cache_read": 0.03 + } + }, + "volcengine/doubao-seed-1.8": { + "id": "volcengine/doubao-seed-1.8", + "name": "Doubao-Seed-1.8", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-18", + "last_updated": "2025-12-18", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.11, + "output": 0.28, + "cache_read": 0.02, + "cache_write": 0.0024 + } + }, + "volcengine/doubao-seed-2.0-mini": { + "id": "volcengine/doubao-seed-2.0-mini", + "name": "Doubao-Seed-2.0-mini", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-02-14", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.03, + "output": 0.28, + "cache_read": 0.01, + "cache_write": 0.0024 + } + }, + "volcengine/doubao-seed-2.0-lite": { + "id": "volcengine/doubao-seed-2.0-lite", + "name": "Doubao-Seed-2.0-lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-02-14", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.09, + "output": 0.51, + "cache_read": 0.02, + "cache_write": 0.0024 + } + }, + "volcengine/doubao-seed-2.0-code": { + "id": "volcengine/doubao-seed-2.0-code", + "name": "Doubao Seed 2.0 Code", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 32000 + }, + "cost": { + "input": 0.9, + "output": 4.48 + } + }, + "anthropic/claude-sonnet-5-free": { + "id": "anthropic/claude-sonnet-5-free", + "name": "Claude Sonnet 5 (Free)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "anthropic/claude-3.5-haiku": { + "id": "anthropic/claude-3.5-haiku", + "name": "Claude 3.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2024-11-04", + "last_updated": "2024-11-04", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 + } + }, + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["image", "text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "anthropic/claude-sonnet-5": { + "id": "anthropic/claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 4 + } + }, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-3.7-sonnet": { + "id": "anthropic/claude-3.7-sonnet", + "name": "Claude 3.7 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-02-24", + "last_updated": "2025-02-24", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-fable-5": { + "id": "anthropic/claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 + } + }, + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["image", "text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", + "modalities": { + "input": ["pdf", "image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-18", + "last_updated": "2026-02-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["image", "text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-06", + "last_updated": "2026-02-06", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "tencent/hy3-preview": { + "id": "tencent/hy3-preview", + "name": "Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.172, + "output": 0.572, + "cache_read": 0.058, + "cache_write": 0 + } + }, + "stepfun/step-3": { + "id": "stepfun/step-3", + "name": "Step-3", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 65536, + "output": 64000 + }, + "cost": { + "input": 0.21, + "output": 0.57 + } + }, + "stepfun/step-3.7-flash": { + "id": "stepfun/step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "input": 256000, + "output": 256000 + }, + "cost": { + "input": 0.2, + "output": 1.15 + } + }, + "stepfun/step-3.7-flash-free": { + "id": "stepfun/step-3.7-flash-free", + "name": "Step 3.7 Flash (Free)", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "input": 256000, + "output": 256000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "stepfun/step-3.5-flash": { + "id": "stepfun/step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-02-02", + "last_updated": "2026-02-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0.1, + "output": 0.3 + } + }, + "kuaishou/kat-coder-pro-v2": { + "id": "kuaishou/kat-coder-pro-v2", + "name": "KAT-Coder-Pro-V2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-03-30", + "last_updated": "2026-03-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 80000 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 + } + }, + "qwen/qwen3-coder-plus": { + "id": "qwen/qwen3-coder-plus", + "name": "Qwen3-Coder-Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "qwen/qwen3.7-plus": { + "id": "qwen/qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.08, + "cache_write": 0.5, + "tiers": [ + { + "input": 1.2, + "output": 4.8, + "cache_read": 0.24, + "cache_write": 1.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.2, + "output": 4.8, + "cache_read": 0.24, + "cache_write": 1.5 + } + } + }, + "qwen/qwen3.7-max": { + "id": "qwen/qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 + } + }, + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen3-Max-Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-01-23", + "last_updated": "2026-01-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 1.2, + "output": 6 + } + }, + "qwen/qwen3.5-flash": { + "id": "qwen/qwen3.5-flash", + "name": "Qwen3.5 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1020000, + "output": 1020000 + }, + "cost": { + "input": 0.1, + "output": 0.4 + } + }, + "qwen/qwen3.5-plus": { + "id": "qwen/qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0.8, + "output": 4.8 + } + }, + "qwen/qwen3.6-plus": { + "id": "qwen/qwen3.6-plus", + "name": "Qwen3.6-Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-30", + "last_updated": "2026-03-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } + } + }, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 + } + }, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 + } + }, + "deepseek/deepseek-v3.2-exp": { + "id": "deepseek/deepseek-v3.2-exp", + "name": "DeepSeek-V3.2-Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 163000, + "output": 64000 + }, + "cost": { + "input": 0.22, + "output": 0.33 + } + }, + "deepseek/deepseek-chat": { + "id": "deepseek/deepseek-chat", + "name": "DeepSeek-V3.2 (Non-thinking Mode)", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.28, + "output": 0.42, + "cache_read": 0.03 + } + }, + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-05", + "last_updated": "2025-12-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.28, + "output": 0.43 + } + }, + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax M2.7 highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 131070 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.611, + "output": 2.4439 + } + }, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-02-13", + "last_updated": "2026-02-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.5-lightning": { + "id": "minimax/minimax-m2.5-lightning", + "name": "MiniMax M2.5 highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-02-13", + "last_updated": "2026-02-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.6, + "output": 4.8, + "cache_read": 0.06, + "cache_write": 0.75 + } + }, + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.38 + } + }, + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.6, + "output": 2.4 + } + }, + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.38 + } + }, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-03-20", + "last_updated": "2026-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 131070 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://zenmux.ai/api/anthropic/v1" + }, + "cost": { + "input": 0.3055, + "output": 1.2219 + } + } + } + }, + "kenari": { + "id": "kenari", + "env": ["KENARI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://kenari.id/v1", + "name": "Kenari", + "doc": "https://kenari.id/docs", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "kimi-k2-6": { + "id": "kimi-k2-6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "glm-5-1": { + "id": "glm-5-1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gemma-4-31b-it": { + "id": "gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "grok-4-3": { + "id": "grok-4-3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 30000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "qwen3-7-plus": { + "id": "qwen3-7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "kimi-k2-7-code": { + "id": "kimi-k2-7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek-v4-flash:free": { + "id": "deepseek-v4-flash:free", + "name": "DeepSeek V4 Flash (Free)", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gpt-5-4-mini": { + "id": "gpt-5-4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2-5": { + "id": "mimo-v2-5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gpt-5-5": { + "id": "gpt-5-5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "glm-5-2": { + "id": "glm-5-2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "grok-build-0-1": { + "id": "grok-build-0-1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2-5-pro": { + "id": "mimo-v2-5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek-v4-pro:free": { + "id": "deepseek-v4-pro:free", + "name": "DeepSeek V4 Pro (Free)", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "gpt-image-2": { + "id": "gpt-image-2", + "name": "GPT-Image-2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 272000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + } + } + }, + "model-oracle-ai": { + "id": "model-oracle-ai", + "env": ["MODEL_ORACLE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.modeloracle.com/api/v1", + "name": "Model Oracle AI", + "doc": "https://modeloracle.com/setup/", + "models": { + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "claude-haiku-4.5": { + "id": "claude-haiku-4.5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + } + }, + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + } + }, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + } + }, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + } + }, + "claude-opus-4.8": { + "id": "claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + } + }, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + } + }, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + } + }, + "auto": { + "id": "auto", + "name": "Auto", + "description": "Model Oracle AI decision engine that selects and routes among configured coding-agent models", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-29", + "last_updated": "2026-07-07", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + } + } + } + }, + "openai": { + "id": "openai", + "env": ["OPENAI_API_KEY"], + "npm": "@ai-sdk/openai", + "name": "OpenAI", + "doc": "https://platform.openai.com/docs/models", + "models": { + "o3": { + "id": "o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "text-embedding-3-large": { + "id": "text-embedding-3-large", + "name": "text-embedding-3-large", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2024-01", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8191, + "output": 3072 + }, + "cost": { + "input": 0.13, + "output": 0 + } + }, + "gpt-5.2-pro": { + "id": "gpt-5.2-pro", + "name": "GPT-5.2 Pro", + "description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 21, + "output": 168 + } + }, + "gpt-5.6": { + "id": "gpt-5.6", + "name": "GPT-5.6", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 10, + "output": 60, + "cache_read": 1, + "cache_write": 12.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + }, + "pro": { + "provider": { + "body": { + "reasoning": { + "mode": "pro" + } + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5 + } + } + }, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-3.5-turbo": { + "id": "gpt-3.5-turbo", + "name": "GPT-3.5-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2021-09-01", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 16385, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 1.5, + "cache_read": 0 + } + }, + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 272000 + }, + "cost": { + "input": 15, + "output": 120 + } + }, + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "gpt-4": { + "id": "gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 30, + "output": 60 + } + }, + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 + } + }, + "o3-pro": { + "id": "o3-pro", + "name": "o3-pro", + "description": "High-effort o3 tier for difficult technical reasoning and careful answers", + "family": "o-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-06-10", + "last_updated": "2025-06-10", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 20, + "output": 80 + } + }, + "chatgpt-image-latest": { + "id": "chatgpt-image-latest", + "name": "chatgpt-image-latest", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "input": 0, + "output": 0 + } + }, + "gpt-4o-2024-05-13": { + "id": "gpt-4o-2024-05-13", + "name": "GPT-4o (2024-05-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 5, + "output": 15 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 + } + }, + "gpt-5-chat-latest": { + "id": "gpt-5-chat-latest", + "name": "GPT-5 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.3-codex-spark": { + "id": "gpt-5.3-codex-spark", + "name": "GPT-5.3 Codex Spark", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex-spark", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 100000, + "output": 32000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.3-chat-latest": { + "id": "gpt-5.3-chat-latest", + "name": "GPT-5.3 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "gpt-4o-2024-08-06": { + "id": "gpt-4o-2024-08-06", + "name": "GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-08-06", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "text-embedding-ada-002": { + "id": "text-embedding-ada-002", + "name": "text-embedding-ada-002", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2022-12", + "release_date": "2022-12-15", + "last_updated": "2022-12-15", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 1536 + }, + "cost": { + "input": 0.1, + "output": 0 + } + }, + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "text-embedding-3-small": { + "id": "text-embedding-3-small", + "name": "text-embedding-3-small", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2024-01", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8191, + "output": 1536 + }, + "cost": { + "input": 0.02, + "output": 0 + } + }, + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "cache_write": 2.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + }, + "pro": { + "provider": { + "body": { + "reasoning": { + "mode": "pro" + } + } + } + } + } + }, + "cost": { + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25, + "tiers": [ + { + "input": 2, + "output": 9, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 9, + "cache_read": 0.2, + "cache_write": 2.5 + } + } + }, + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "gpt-realtime-2.1": { + "id": "gpt-realtime-2.1", + "name": "GPT-Realtime-2.1", + "description": "Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2026-07-06", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text", "audio", "image"], + "output": ["text", "audio"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 96000, + "output": 32000 + }, + "cost": { + "input": 4, + "output": 24, + "cache_read": 0.4, + "input_audio": 32, + "output_audio": 64 + } + }, + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + }, + "pro": { + "provider": { + "body": { + "reasoning": { + "mode": "pro" + } + } + } + } + } + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 6.25, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 6.25 + } + } + }, + "gpt-5.1-chat-latest": { + "id": "gpt-5.1-chat-latest", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT-5.1 for polished assistants, writing, and product conversations", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.2-chat-latest": { + "id": "gpt-5.2-chat-latest", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "o4-mini-deep-research": { + "id": "o4-mini-deep-research", + "name": "o4-mini-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-06-26", + "last_updated": "2024-06-26", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "gpt-image-1.5": { + "id": "gpt-image-1.5", + "name": "gpt-image-1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-11-25", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "input": 0, + "output": 0 + } + }, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "gpt-4o-2024-11-20": { + "id": "gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-11-20", + "last_updated": "2024-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "o1": { + "id": "o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 15, + "output": 60, + "cache_read": 7.5 + } + }, + "o1-pro": { + "id": "o1-pro", + "name": "o1-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2025-03-19", + "last_updated": "2025-03-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 150, + "output": 600 + } + }, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } + } + }, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "o3-deep-research": { + "id": "o3-deep-research", + "name": "o3-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-06-26", + "last_updated": "2024-06-26", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 10, + "output": 40, + "cache_read": 2.5 + } + }, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "gpt-image-1": { + "id": "gpt-image-1", + "name": "gpt-image-1", + "description": "OpenAI image model for production generation, edits, and brand-safe visual workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-04-24", + "last_updated": "2025-04-24", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "input": 0, + "output": 0 + } + }, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 + } + }, + "gpt-4-turbo": { + "id": "gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 10, + "output": 30 + } + }, + "gpt-image-1-mini": { + "id": "gpt-image-1-mini", + "name": "gpt-image-1-mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-09-26", + "last_updated": "2025-09-26", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "input": 0, + "output": 0 + } + }, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 + } + }, + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "gpt-5.5-pro": { + "id": "gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 10, + "output": 60, + "cache_read": 1, + "cache_write": 12.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + }, + "pro": { + "provider": { + "body": { + "reasoning": { + "mode": "pro" + } + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5 + } + } + }, + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "gpt-image-2": { + "id": "gpt-image-2", + "name": "gpt-image-2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "input": 0, + "output": 0 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 1.25 + } + }, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + } + } + }, + "berget": { + "id": "berget", + "env": ["BERGET_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.berget.ai/v1", + "name": "Berget.AI", + "doc": "https://api.berget.ai", + "models": { + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2025-04-27", + "last_updated": "2025-04-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.99, + "output": 0.99 + } + }, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.83, + "output": 3.85, + "cache_read": 0.16 + } + }, + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-12", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["audio", "image", "text", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.275, + "output": 0.55 + } + }, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT-OSS-120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.22, + "output": 0.83 + } + }, + "mistralai/Mistral-Medium-3.5-128B": { + "id": "mistralai/Mistral-Medium-3.5-128B", + "name": "Mistral Medium 3.5 128B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2026-04", + "release_date": "2026-04-29", + "last_updated": "2026-04-29", + "modalities": { + "input": ["image", "text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 1.65, + "output": 5.5 + } + }, + "mistralai/Mistral-Small-3.2-24B-Instruct-2506": { + "id": "mistralai/Mistral-Small-3.2-24B-Instruct-2506", + "name": "Mistral Small 3.2 24B Instruct 2506", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-10-01", + "last_updated": "2025-10-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32000, + "output": 8192 + }, + "cost": { + "input": 0.33, + "output": 0.33 + } + }, + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-12", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.77, + "output": 2.75 + } + }, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 524288, + "output": 32768 + }, + "cost": { + "input": 1.54, + "output": 4.84 + } + } + } + }, + "snowflake-cortex": { + "id": "snowflake-cortex", + "env": ["SNOWFLAKE_ACCOUNT", "SNOWFLAKE_CORTEX_PAT"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://${SNOWFLAKE_ACCOUNT}.snowflakecomputing.com/api/v2/cortex/v1", + "name": "Snowflake Cortex", + "doc": "https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-rest-api", + "models": { + "openai-gpt-5.1": { + "id": "openai-gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "snowflake-llama3.3-70b": { + "id": "snowflake-llama3.3-70b", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + } + }, + "openai-gpt-5.2": { + "id": "openai-gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + } + }, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 16384 + } + }, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "status": "beta", + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + } + }, + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek-R1", + "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + } + }, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "openai-gpt-5": { + "id": "openai-gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "status": "beta" + }, + "openai-gpt-5.5": { + "id": "openai-gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "status": "beta" + }, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + } + }, + "openai-gpt-5-nano": { + "id": "openai-gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "status": "beta" + }, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 16384 + } + }, + "mistral-large2": { + "id": "mistral-large2", + "name": "Mistral Large (latest)", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + } + }, + "openai-gpt-4.1": { + "id": "openai-gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + } + }, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 16384 + } + }, + "openai-gpt-5.4": { + "id": "openai-gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "status": "beta", + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + } + }, + "gemini-3.1-pro": { + "id": "gemini-3.1-pro", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + } + }, + "openai-gpt-5-mini": { + "id": "openai-gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 272000, + "input": 272000, + "output": 8192 + }, + "status": "beta" + } + } + }, + "tencent-token-plan": { + "id": "tencent-token-plan", + "env": ["TENCENT_TOKEN_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.lkeap.cloud.tencent.com/plan/v3", + "name": "Tencent Token Plan", + "doc": "https://cloud.tencent.com/document/product/1823/130060", + "models": { + "hy3": { + "id": "hy3", + "name": "Hy3", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-07-06", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + } + } + }, + "github-models": { + "id": "github-models", + "env": ["GITHUB_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://models.github.ai/inference", + "name": "GitHub Models", + "doc": "https://docs.github.com/en/github-models", + "models": { + "ai21-labs/ai21-jamba-1.5-mini": { + "id": "ai21-labs/ai21-jamba-1.5-mini", + "name": "AI21 Jamba 1.5 Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "jamba", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-08-29", + "last_updated": "2024-08-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "ai21-labs/ai21-jamba-1.5-large": { + "id": "ai21-labs/ai21-jamba-1.5-large", + "name": "AI21 Jamba 1.5 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "jamba", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-08-29", + "last_updated": "2024-08-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "core42/jais-30b-chat": { + "id": "core42/jais-30b-chat", + "name": "JAIS 30b Chat", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "jais", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-03", + "release_date": "2023-08-30", + "last_updated": "2023-08-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "xai/grok-3-mini": { + "id": "xai/grok-3-mini", + "name": "Grok 3 Mini", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-09", + "last_updated": "2024-12-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "xai/grok-3": { + "id": "xai/grok-3", + "name": "Grok 3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-09", + "last_updated": "2024-12-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3.5-moe-instruct": { + "id": "microsoft/phi-3.5-moe-instruct", + "name": "Phi-3.5-MoE instruct (128k)", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-small-128k-instruct": { + "id": "microsoft/phi-3-small-128k-instruct", + "name": "Phi-3-small instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3.5-mini-instruct": { + "id": "microsoft/phi-3.5-mini-instruct", + "name": "Phi-3.5-mini instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-medium-128k-instruct": { + "id": "microsoft/phi-3-medium-128k-instruct", + "name": "Phi-3-medium instruct (128k)", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-small-8k-instruct": { + "id": "microsoft/phi-3-small-8k-instruct", + "name": "Phi-3-small instruct (8k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-4-reasoning": { + "id": "microsoft/phi-4-reasoning", + "name": "Phi-4-Reasoning", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/mai-ds-r1": { + "id": "microsoft/mai-ds-r1", + "name": "MAI-DS-R1", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "mai", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 65536, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-4-mini-instruct": { + "id": "microsoft/phi-4-mini-instruct", + "name": "Phi-4-mini-instruct", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-4": { + "id": "microsoft/phi-4", + "name": "Phi-4", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 16000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3.5-vision-instruct": { + "id": "microsoft/phi-3.5-vision-instruct", + "name": "Phi-3.5-vision instruct (128k)", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-medium-4k-instruct": { + "id": "microsoft/phi-3-medium-4k-instruct", + "name": "Phi-3-medium instruct (4k)", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 4096, + "output": 1024 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-4-multimodal-instruct": { + "id": "microsoft/phi-4-multimodal-instruct", + "name": "Phi-4-multimodal-instruct", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-4-mini-reasoning": { + "id": "microsoft/phi-4-mini-reasoning", + "name": "Phi-4-mini-reasoning", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-mini-128k-instruct": { + "id": "microsoft/phi-3-mini-128k-instruct", + "name": "Phi-3-mini instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "microsoft/phi-3-mini-4k-instruct": { + "id": "microsoft/phi-3-mini-4k-instruct", + "name": "Phi-3-mini instruct (4k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 4096, + "output": 1024 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o3": { + "id": "openai/o3", + "name": "OpenAI o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o1-mini": { + "id": "openai/o1-mini", + "name": "OpenAI o1-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2023-10", + "release_date": "2024-09-12", + "last_updated": "2024-12-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "OpenAI o4-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o1-preview": { + "id": "openai/o1-preview", + "name": "OpenAI o1-preview", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2023-10", + "release_date": "2024-09-12", + "last_updated": "2024-09-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "OpenAI o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/o1": { + "id": "openai/o1", + "name": "OpenAI o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2023-10", + "release_date": "2024-09-12", + "last_updated": "2024-12-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/mistral-small-2503": { + "id": "mistral-ai/mistral-small-2503", + "name": "Mistral Small 3.1", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2025-03-01", + "last_updated": "2025-03-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/mistral-nemo": { + "id": "mistral-ai/mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/mistral-medium-2505": { + "id": "mistral-ai/mistral-medium-2505", + "name": "Mistral Medium 3 (25.05)", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2025-05-01", + "last_updated": "2025-05-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/mistral-large-2411": { + "id": "mistral-ai/mistral-large-2411", + "name": "Mistral Large 24.11", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/ministral-3b": { + "id": "mistral-ai/ministral-3b", + "name": "Ministral 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mistral-ai/codestral-2501": { + "id": "mistral-ai/codestral-2501", + "name": "Codestral 25.01", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "codestral", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "cohere/cohere-command-a": { + "id": "cohere/cohere-command-a", + "name": "Cohere Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "cohere/cohere-command-r-08-2024": { + "id": "cohere/cohere-command-r-08-2024", + "name": "Cohere Command R 08-2024", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-08-01", + "last_updated": "2024-08-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "cohere/cohere-command-r-plus": { + "id": "cohere/cohere-command-r-plus", + "name": "Cohere Command R+", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-04-04", + "last_updated": "2024-08-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "cohere/cohere-command-r-plus-08-2024": { + "id": "cohere/cohere-command-r-plus-08-2024", + "name": "Cohere Command R+ 08-2024", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-08-01", + "last_updated": "2024-08-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "cohere/cohere-command-r": { + "id": "cohere/cohere-command-r", + "name": "Cohere Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-03-11", + "last_updated": "2024-08-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/meta-llama-3-8b-instruct": { + "id": "meta/meta-llama-3-8b-instruct", + "name": "Meta-Llama-3-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/llama-3.2-11b-vision-instruct": { + "id": "meta/llama-3.2-11b-vision-instruct", + "name": "Llama-3.2-11B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/meta-llama-3.1-405b-instruct": { + "id": "meta/meta-llama-3.1-405b-instruct", + "name": "Meta-Llama-3.1-405B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/llama-4-scout-17b-16e-instruct": { + "id": "meta/llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/llama-3.3-70b-instruct": { + "id": "meta/llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/meta-llama-3.1-70b-instruct": { + "id": "meta/meta-llama-3.1-70b-instruct", + "name": "Meta-Llama-3.1-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/meta-llama-3.1-8b-instruct": { + "id": "meta/meta-llama-3.1-8b-instruct", + "name": "Meta-Llama-3.1-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/llama-3.2-90b-vision-instruct": { + "id": "meta/llama-3.2-90b-vision-instruct", + "name": "Llama-3.2-90B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/llama-4-maverick-17b-128e-instruct-fp8": { + "id": "meta/llama-4-maverick-17b-128e-instruct-fp8", + "name": "Llama 4 Maverick 17B 128E Instruct FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "meta/meta-llama-3-70b-instruct": { + "id": "meta/meta-llama-3-70b-instruct", + "name": "Meta-Llama-3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek/deepseek-r1-0528": { + "id": "deepseek/deepseek-r1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 65536, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek/deepseek-v3-0324": { + "id": "deepseek/deepseek-v3-0324", + "name": "DeepSeek-V3-0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek/deepseek-r1": { + "id": "deepseek/deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 65536, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + } + } + }, + "neuralwatt": { + "id": "neuralwatt", + "env": ["NEURALWATT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.neuralwatt.com/v1", + "name": "Neuralwatt", + "doc": "https://portal.neuralwatt.com/docs", + "models": { + "kimi-k2.5-fast": { + "id": "kimi-k2.5-fast", + "name": "Kimi K2.5 Fast", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.52, + "output": 2.59, + "cache_read": 0.13 + } + }, + "kimi-k2.6-flex": { + "id": "kimi-k2.6-flex", + "name": "Kimi K2.6 Flex", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.345, + "output": 1.61, + "cache_read": 0.08625 + } + }, + "glm-5.2-short-fast-flex": { + "id": "glm-5.2-short-fast-flex", + "name": "GLM 5.2 Short Fast Flex", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 199984, + "output": 199984 + }, + "cost": { + "input": 0.725, + "output": 2.25, + "cache_read": 0.18125 + } + }, + "glm-5.2-flex": { + "id": "glm-5.2-flex", + "name": "GLM 5.2 Flex", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048560, + "output": 1048560 + }, + "cost": { + "input": 0.725, + "output": 2.25, + "cache_read": 0.18125 + } + }, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM 5.2", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048560, + "output": 1048560 + }, + "cost": { + "input": 1.45, + "output": 4.5, + "cache_read": 0.3625 + } + }, + "glm-5.2-short-fast": { + "id": "glm-5.2-short-fast", + "name": "GLM 5.2 Short Fast", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 199984, + "output": 199984 + }, + "cost": { + "input": 1.45, + "output": 4.5, + "cache_read": 0.3625 + } + }, + "qwen3.5-397b-fast": { + "id": "qwen3.5-397b-fast", + "name": "Qwen3.5 397B Fast", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.69, + "output": 4.14, + "cache_read": 0.1725 + } + }, + "kimi-k2.6-fast": { + "id": "kimi-k2.6-fast", + "name": "Kimi K2.6 Fast", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.69, + "output": 3.22, + "cache_read": 0.1725 + } + }, + "qwen3.6-35b-fast": { + "id": "qwen3.6-35b-fast", + "name": "Qwen3.6 35B Fast", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "qwen3.6", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131056, + "output": 131056 + }, + "cost": { + "input": 0.29, + "output": 1.15, + "cache_read": 0.0725 + } + }, + "glm-5.2-short-flex": { + "id": "glm-5.2-short-flex", + "name": "GLM 5.2 Short Flex", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 199984, + "output": 199984 + }, + "cost": { + "input": 0.725, + "output": 2.25, + "cache_read": 0.18125 + } + }, + "glm-5.2-fast": { + "id": "glm-5.2-fast", + "name": "GLM 5.2 Fast", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048560, + "output": 1048560 + }, + "cost": { + "input": 1.45, + "output": 4.5, + "cache_read": 0.3625 + } + }, + "glm-5.2-short": { + "id": "glm-5.2-short", + "name": "GLM 5.2 Short", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-17", + "last_updated": "2026-06-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 199984, + "output": 199984 + }, + "cost": { + "input": 1.45, + "output": 4.5, + "cache_read": 0.3625 + } + }, + "kimi-k2.7-code-flex": { + "id": "kimi-k2.7-code-flex", + "name": "Kimi K2.7 Code Flex", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.475, + "output": 2, + "cache_read": 0.11875 + } + }, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.69, + "output": 3.22, + "cache_read": 0.1725 + } + }, + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.52, + "output": 2.59, + "cache_read": 0.13 + } + }, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.2375 + } + }, + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131056, + "output": 131056 + }, + "cost": { + "input": 0.29, + "output": 1.15, + "cache_read": 0.0725 + } + }, + "Qwen/Qwen3.5-397B-A17B-FP8": { + "id": "Qwen/Qwen3.5-397B-A17B-FP8", + "name": "Qwen3.5 397B A17B FP8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-01", + "last_updated": "2026-02-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262128, + "output": 262128 + }, + "cost": { + "input": 0.69, + "output": 4.14, + "cache_read": 0.1725 + } + } + } + }, + "siliconflow-cn": { + "id": "siliconflow-cn", + "env": ["SILICONFLOW_CN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.siliconflow.cn/v1", + "name": "SiliconFlow (China)", + "doc": "https://cloud.siliconflow.com/models", + "models": { + "baidu/ERNIE-4.5-300B-A47B": { + "id": "baidu/ERNIE-4.5-300B-A47B", + "name": "baidu/ERNIE-4.5-300B-A47B", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "ernie", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-02", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.28, + "output": 1.1 + } + }, + "ByteDance-Seed/Seed-OSS-36B-Instruct": { + "id": "ByteDance-Seed/Seed-OSS-36B-Instruct", + "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "seed", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-04", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.21, + "output": 0.57 + } + }, + "stepfun-ai/Step-3.5-Flash": { + "id": "stepfun-ai/Step-3.5-Flash", + "name": "stepfun-ai/Step-3.5-Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "family": "step", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.1, + "output": 0.3 + } + }, + "inclusionAI/Ling-flash-2.0": { + "id": "inclusionAI/Ling-flash-2.0", + "name": "inclusionAI/Ling-flash-2.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "ling", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-18", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.14, + "output": 0.57 + } + }, + "Pro/moonshotai/Kimi-K2.6": { + "id": "Pro/moonshotai/Kimi-K2.6", + "name": "Pro/moonshotai/Kimi-K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.16 + } + }, + "Pro/moonshotai/Kimi-K2.5": { + "id": "Pro/moonshotai/Kimi-K2.5", + "name": "Pro/moonshotai/Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.45, + "output": 2.25 + } + }, + "Pro/zai-org/GLM-5": { + "id": "Pro/zai-org/GLM-5", + "name": "Pro/zai-org/GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 205000, + "output": 205000 + }, + "cost": { + "input": 1, + "output": 3.2 + } + }, + "Pro/zai-org/GLM-5.1": { + "id": "Pro/zai-org/GLM-5.1", + "name": "Pro/zai-org/GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-08", + "last_updated": "2026-04-08", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 205000, + "output": 205000 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_write": 0 + } + }, + "Pro/deepseek-ai/DeepSeek-R1": { + "id": "Pro/deepseek-ai/DeepSeek-R1", + "name": "Pro/deepseek-ai/DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-05-28", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.5, + "output": 2.18 + } + }, + "Pro/deepseek-ai/DeepSeek-V3.1-Terminus": { + "id": "Pro/deepseek-ai/DeepSeek-V3.1-Terminus", + "name": "Pro/deepseek-ai/DeepSeek-V3.1-Terminus", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-29", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.27, + "output": 1 + } + }, + "Pro/deepseek-ai/DeepSeek-V3.2": { + "id": "Pro/deepseek-ai/DeepSeek-V3.2", + "name": "Pro/deepseek-ai/DeepSeek-V3.2", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-12-03", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.27, + "output": 0.42 + } + }, + "Pro/deepseek-ai/DeepSeek-V3": { + "id": "Pro/deepseek-ai/DeepSeek-V3", + "name": "Pro/deepseek-ai/DeepSeek-V3", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-12-26", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.25, + "output": 1 + } + }, + "Pro/MiniMaxAI/MiniMax-M2.5": { + "id": "Pro/MiniMaxAI/MiniMax-M2.5", + "name": "Pro/MiniMaxAI/MiniMax-M2.5", + "description": "Frontier MiniMax model for engineering, office tasks, and agentic reasoning", + "family": "minimax", + "attachment": false, + "reasoning": false, + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-02-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 192000, + "output": 131000 + }, + "cost": { + "input": 0.3, + "output": 1.22 + } + }, + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen/Qwen3.6-35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.23, + "output": 1.86 + } + }, + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen/Qwen3.5-397B-A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.29, + "output": 1.74 + } + }, + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.13, + "output": 0.6 + } + }, + "Qwen/Qwen3.5-122B-A10B": { + "id": "Qwen/Qwen3.5-122B-A10B", + "name": "Qwen/Qwen3.5-122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.29, + "output": 2.32 + } + }, + "Qwen/Qwen3.5-27B": { + "id": "Qwen/Qwen3.5-27B", + "name": "Qwen/Qwen3.5-27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-25", + "last_updated": "2026-02-25", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.26, + "output": 2.09 + } + }, + "Qwen/Qwen3-8B": { + "id": "Qwen/Qwen3-8B", + "name": "Qwen/Qwen3-8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.06, + "output": 0.06 + } + }, + "Qwen/Qwen3.5-4B": { + "id": "Qwen/Qwen3.5-4B", + "name": "Qwen/Qwen3.5-4B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen/Qwen3.5-9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.22, + "output": 1.74 + } + }, + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen/Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.14, + "output": 0.57 + } + }, + "Qwen/Qwen3-14B": { + "id": "Qwen/Qwen3-14B", + "name": "Qwen/Qwen3-14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.07, + "output": 0.28 + } + }, + "Qwen/Qwen3.5-35B-A3B": { + "id": "Qwen/Qwen3.5-35B-A3B", + "name": "Qwen/Qwen3.5-35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-25", + "last_updated": "2026-02-25", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.23, + "output": 1.86 + } + }, + "Qwen/Qwen3-VL-32B-Thinking": { + "id": "Qwen/Qwen3-VL-32B-Thinking", + "name": "Qwen/Qwen3-VL-32B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-21", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.2, + "output": 1.5 + } + }, + "Qwen/Qwen3-VL-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "name": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-05", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.29, + "output": 1 + } + }, + "Qwen/Qwen2.5-72B-Instruct": { + "id": "Qwen/Qwen2.5-72B-Instruct", + "name": "Qwen/Qwen2.5-72B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 33000, + "output": 4000 + }, + "cost": { + "input": 0.59, + "output": 0.59 + } + }, + "Qwen/Qwen3-VL-32B-Instruct": { + "id": "Qwen/Qwen3-VL-32B-Instruct", + "name": "Qwen/Qwen3-VL-32B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-21", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.2, + "output": 0.6 + } + }, + "Qwen/Qwen3-VL-235B-A22B-Thinking": { + "id": "Qwen/Qwen3-VL-235B-A22B-Thinking", + "name": "Qwen/Qwen3-VL-235B-A22B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-04", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.45, + "output": 3.5 + } + }, + "Qwen/Qwen3-VL-8B-Instruct": { + "id": "Qwen/Qwen3-VL-8B-Instruct", + "name": "Qwen/Qwen3-VL-8B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-15", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.18, + "output": 0.68 + } + }, + "Qwen/Qwen3-VL-30B-A3B-Thinking": { + "id": "Qwen/Qwen3-VL-30B-A3B-Thinking", + "name": "Qwen/Qwen3-VL-30B-A3B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-11", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.29, + "output": 1 + } + }, + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-30", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.09, + "output": 0.3 + } + }, + "Qwen/Qwen3-Coder-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "name": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-01", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.07, + "output": 0.28 + } + }, + "Qwen/Qwen3-Coder-480B-A35B-Instruct": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "name": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-31", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.25, + "output": 1 + } + }, + "Qwen/Qwen3-VL-235B-A22B-Instruct": { + "id": "Qwen/Qwen3-VL-235B-A22B-Instruct", + "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-04", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 262000 + }, + "cost": { + "input": 0.3, + "output": 1.5 + } + }, + "Qwen/Qwen2.5-7B-Instruct": { + "id": "Qwen/Qwen2.5-7B-Instruct", + "name": "Qwen/Qwen2.5-7B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 33000, + "output": 4000 + }, + "cost": { + "input": 0.05, + "output": 0.05 + } + }, + "PaddlePaddle/PaddleOCR-VL-1.5": { + "id": "PaddlePaddle/PaddleOCR-VL-1.5", + "name": "PaddlePaddle/PaddleOCR-VL-1.5", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 16384, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "tencent/Hunyuan-A13B-Instruct": { + "id": "tencent/Hunyuan-A13B-Instruct", + "name": "tencent/Hunyuan-A13B-Instruct", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.14, + "output": 0.57 + } + }, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1049000, + "output": 262000 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_write": 0 + } + }, + "zai-org/GLM-4.5-Air": { + "id": "zai-org/GLM-4.5-Air", + "name": "zai-org/GLM-4.5-Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131000, + "output": 131000 + }, + "cost": { + "input": 0.14, + "output": 0.86 + } + }, + "deepseek-ai/DeepSeek-R1": { + "id": "deepseek-ai/DeepSeek-R1", + "name": "deepseek-ai/DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-05-28", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.5, + "output": 2.18 + } + }, + "deepseek-ai/DeepSeek-V3.1-Terminus": { + "id": "deepseek-ai/DeepSeek-V3.1-Terminus", + "name": "deepseek-ai/DeepSeek-V3.1-Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-29", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.27, + "output": 1 + } + }, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.003 + } + }, + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "deepseek-ai/DeepSeek-V4-Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1049000, + "output": 393000 + }, + "cost": { + "input": 1.74, + "output": 3.48, + "cache_read": 0.145 + } + }, + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "deepseek-ai/DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2025-12-03", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.27, + "output": 0.42 + } + }, + "deepseek-ai/DeepSeek-OCR": { + "id": "deepseek-ai/DeepSeek-OCR", + "name": "deepseek-ai/DeepSeek-OCR", + "description": "OCR model for extracting structured text from documents and screenshots", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-10-20", + "last_updated": "2025-10-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "deepseek-ai/DeepSeek-V3": { + "id": "deepseek-ai/DeepSeek-V3", + "name": "deepseek-ai/DeepSeek-V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-12-26", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 164000, + "output": 164000 + }, + "cost": { + "input": 0.25, + "output": 1 + } + } + } + }, + "merge-gateway": { + "id": "merge-gateway", + "env": ["MERGE_GATEWAY_API_KEY"], + "npm": "merge-gateway-ai-sdk-provider", + "name": "Merge Gateway", + "doc": "https://docs.merge.dev/merge-gateway", + "models": { + "xai/grok-4.3": { + "id": "xai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 30000 + }, + "cost": { + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } + } + }, + "xai/grok-4.20-0309-reasoning": { + "id": "xai/grok-4.20-0309-reasoning", + "name": "Grok 4.20 (Reasoning)", + "description": "Reasoning Grok for document-heavy analysis and long-horizon tool use", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-09", + "last_updated": "2026-03-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 30000 + }, + "cost": { + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } + } + }, + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4 + } + }, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.6, + "output": 2.5 + } + }, + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.6, + "output": 3 + } + }, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4 + } + }, + "moonshotai/kimi-k2.7-code-highspeed": { + "id": "moonshotai/kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code Highspeed", + "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 1.9, + "output": 8 + } + }, + "mistral/codestral-latest": { + "id": "mistral/codestral-latest", + "name": "Codestral (latest)", + "description": "Mistral code model for completions, refactors, and developer IDE workflows", + "family": "codestral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-05-29", + "last_updated": "2025-01-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 4096 + }, + "cost": { + "input": 0.3, + "output": 0.9 + } + }, + "mistral/mistral-large-latest": { + "id": "mistral/mistral-large-latest", + "name": "Mistral Large (latest)", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.5, + "output": 1.5 + } + }, + "mistral/devstral-small-2507": { + "id": "mistral/devstral-small-2507", + "name": "Devstral Small", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-07-10", + "last_updated": "2025-07-10", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 128000 + }, + "status": "deprecated", + "cost": { + "input": 0.1, + "output": 0.3 + } + }, + "mistral/pixtral-large-latest": { + "id": "mistral/pixtral-large-latest", + "name": "Pixtral Large (latest)", + "description": "Mistral's larger vision model for document-heavy image understanding and chat", + "family": "pixtral", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2024-11-04", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 128000 + }, + "cost": { + "input": 2, + "output": 6 + } + }, + "mistral/mistral-medium-latest": { + "id": "mistral/mistral-medium-latest", + "name": "Mistral Medium (latest)", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.4, + "output": 2 + } + }, + "mistral/mistral-small-latest": { + "id": "mistral/mistral-small-latest", + "name": "Mistral Small (latest)", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 0.15, + "output": 0.6 + } + }, + "mistral/mistral-medium-2505": { + "id": "mistral/mistral-medium-2505", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.4, + "output": 2 + } + }, + "mistral/mistral-large-2411": { + "id": "mistral/mistral-large-2411", + "name": "Mistral Large 2.1", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-18", + "last_updated": "2024-11-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 16384 + }, + "cost": { + "input": 2, + "output": 6 + } + }, + "mistral/magistral-medium-latest": { + "id": "mistral/magistral-medium-latest", + "name": "Magistral Medium (latest)", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-medium", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-03-17", + "last_updated": "2025-03-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2, + "output": 5 + } + }, + "mistral/devstral-medium-latest": { + "id": "mistral/devstral-medium-latest", + "name": "Devstral 2 (latest)", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "status": "deprecated", + "cost": { + "input": 0.4, + "output": 2 + } + }, + "mistral/devstral-2512": { + "id": "mistral/devstral-2512", + "name": "Devstral 2", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "status": "deprecated", + "cost": { + "input": 0.4, + "output": 2 + } + }, + "mistral/mistral-large-2512": { + "id": "mistral/mistral-large-2512", + "name": "Mistral Large 3", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.5, + "output": 1.5 + } + }, + "mistral/devstral-medium-2507": { + "id": "mistral/devstral-medium-2507", + "name": "Devstral Medium", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-07-10", + "last_updated": "2025-07-10", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 128000 + }, + "status": "deprecated", + "cost": { + "input": 0.4, + "output": 2 + } + }, + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + } + }, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 + } + }, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 + } + }, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + } + }, + "google/gemini-3.1-pro-preview-customtools": { + "id": "google/gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemini-flash-lite-latest": { + "id": "google/gemini-flash-lite-latest", + "name": "Gemini Flash-Lite Latest", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 + } + }, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + } + }, + "google/gemini-3-pro-preview": { + "id": "google/gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 + } + }, + "google/gemini-flash-latest": { + "id": "google/gemini-flash-latest", + "name": "Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.075, + "input_audio": 1 + } + }, + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 + } + }, + "openai/gpt-4o-2024-05-13": { + "id": "openai/gpt-4o-2024-05-13", + "name": "GPT-4o (2024-05-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 5, + "output": 15 + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 + } + }, + "openai/gpt-5-chat-latest": { + "id": "openai/gpt-5-chat-latest", + "name": "GPT-5 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.3-chat-latest": { + "id": "openai/gpt-5.3-chat-latest", + "name": "GPT-5.3 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-4o-2024-08-06": { + "id": "openai/gpt-4o-2024-08-06", + "name": "GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-08-06", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-5.1-chat-latest": { + "id": "openai/gpt-5.1-chat-latest", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT-5.1 for polished assistants, writing, and product conversations", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.2-chat-latest": { + "id": "openai/gpt-5.2-chat-latest", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "openai/gpt-4o-2024-11-20": { + "id": "openai/gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-11-20", + "last_updated": "2024-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/o1": { + "id": "openai/o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 15, + "output": 60, + "cache_read": 7.5 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } + } + }, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 + } + }, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 + } + }, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + }, + "zai/glm-4.7": { + "id": "zai/glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "zai/glm-4.5": { + "id": "zai/glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "zai/glm-4.7-flashx": { + "id": "zai/glm-4.7-flashx", + "name": "GLM-4.7-FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0.07, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0 + } + }, + "zai/glm-5.1": { + "id": "zai/glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 + } + }, + "zai/glm-4.6": { + "id": "zai/glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "zai/glm-5.2": { + "id": "zai/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 50000 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4 + } + }, + "zai/glm-4.5-air": { + "id": "zai/glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.2, + "output": 1.1, + "cache_read": 0.03, + "cache_write": 0 + } + }, + "zai/glm-5": { + "id": "zai/glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.2, + "cache_write": 0 + } + }, + "zai/glm-5-turbo": { + "id": "zai/glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.2, + "output": 4, + "cache_read": 0.24, + "cache_write": 0 + } + }, + "anthropic/claude-haiku-4-5-20251001": { + "id": "anthropic/claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "anthropic/claude-opus-4-1-20250805": { + "id": "anthropic/claude-opus-4-1-20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-opus-4-5-20251101": { + "id": "anthropic/claude-opus-4-5-20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-opus-4-8": { + "id": "anthropic/claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 128000 + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25 + } + }, + "anthropic/claude-opus-4-20250514": { + "id": "anthropic/claude-opus-4-20250514", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-sonnet-4-20250514": { + "id": "anthropic/claude-sonnet-4-20250514", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-sonnet-4-5-20250929": { + "id": "anthropic/claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "cohere/command-a-03-2025": { + "id": "cohere/command-a-03-2025", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8000 + }, + "cost": { + "input": 2.5, + "output": 10 + } + }, + "cohere/command-r-08-2024": { + "id": "cohere/command-r-08-2024", + "name": "Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4000 + }, + "cost": { + "input": 0.15, + "output": 0.6 + } + }, + "cohere/command-r7b-12-2024": { + "id": "cohere/command-r7b-12-2024", + "name": "Command R7B", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-12-02", + "last_updated": "2024-12-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4000 + }, + "cost": { + "input": 0.0375, + "output": 0.15 + } + }, + "cohere/command-r-plus-08-2024": { + "id": "cohere/command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 4000 + }, + "cost": { + "input": 2.5, + "output": 10 + } + }, + "alibaba/qwen3.7-max": { + "id": "alibaba/qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 250000 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 1.65, + "output": 4.95 + } + }, + "alibaba/qwen3.6-plus": { + "id": "alibaba/qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 250000 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3 + } + }, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 + } + }, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 + } + }, + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax-M2.1", + "description": "Earlier MiniMax agent model for practical coding and productivity tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2 + } + }, + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 128000 + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512000, + "output": 128000 + }, + "cost": { + "input": 0.6, + "output": 2.4 + } + }, + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax-M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 196608, + "output": 128000 + }, + "cost": { + "input": 0.3, + "output": 1.2 + } + }, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.5-highspeed": { + "id": "minimax/minimax-m2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-02-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 + } + } + } + }, + "qihang-ai": { + "id": "qihang-ai", + "env": ["QIHANG_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.qhaigc.net/v1", + "name": "QiHang", + "doc": "https://www.qhaigc.net/docs", + "models": { + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-10-01", + "last_updated": "2025-10-01", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.14, + "output": 0.71 + } + }, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.09, + "output": 0.71, + "tiers": [ + { + "input": 0.09, + "output": 0.71, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 0.09, + "output": 0.71 + } + } + }, + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 0.71, + "output": 3.57 + } + }, + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2 + } + }, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.43, + "output": 2.14 + } + }, + "gemini-3-pro-preview": { + "id": "gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-19", + "last_updated": "2025-11-19", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65000 + }, + "cost": { + "input": 0.57, + "output": 3.43 + } + }, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5-Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0.04, + "output": 0.29 + } + }, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.07, + "output": 0.43, + "tiers": [ + { + "input": 0.07, + "output": 0.43, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 0.07, + "output": 0.43 + } + } + }, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.14, + "output": 1.14 + } + } + } + }, + "xiaomi-token-plan-ams": { + "id": "xiaomi-token-plan-ams", + "env": ["XIAOMI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://token-plan-ams.xiaomimimo.com/v1", + "name": "Xiaomi Token Plan (Europe)", + "doc": "https://platform.xiaomimimo.com/#/docs", + "models": { + "mimo-v2.5-tts": { + "id": "mimo-v2.5-tts", + "name": "MiMo-V2.5-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0 + } + }, + "mimo-v2-pro": { + "id": "mimo-v2-pro", + "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 131072 + }, + "status": "deprecated", + "cost": { + "input": 0, + "output": 0, + "cache_read": 0 + } + }, + "mimo-v2-tts": { + "id": "mimo-v2-tts", + "name": "MiMo-V2-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2.5": { + "id": "mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0 + } + }, + "mimo-v2.5-tts-voicedesign": { + "id": "mimo-v2.5-tts-voicedesign", + "name": "MiMo-V2.5-TTS-VoiceDesign", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2.5-tts-voiceclone": { + "id": "mimo-v2.5-tts-voiceclone", + "name": "MiMo-V2.5-TTS-VoiceClone", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + } + } + }, + "modelscope": { + "id": "modelscope", + "env": ["MODELSCOPE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api-inference.modelscope.cn/v1", + "name": "ModelScope", + "doc": "https://modelscope.cn/docs/model-service/API-Inference/intro", + "models": { + "Qwen/Qwen3-30B-A3B-Thinking-2507": { + "id": "Qwen/Qwen3-30B-A3B-Thinking-2507", + "name": "Qwen3 30B A3B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen3-235B-A22B-Thinking-2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "Qwen/Qwen3-Coder-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-07-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "ZhipuAI/GLM-4.6": { + "id": "ZhipuAI/GLM-4.6", + "name": "GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 98304 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "ZhipuAI/GLM-4.5": { + "id": "ZhipuAI/GLM-4.5", + "name": "GLM-4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0, + "output": 0 + } + } + } + }, + "groq": { + "id": "groq", + "env": ["GROQ_API_KEY"], + "npm": "@ai-sdk/groq", + "name": "Groq", + "doc": "https://console.groq.com/docs/models", + "models": { + "llama-3.3-70b-versatile": { + "id": "llama-3.3-70b-versatile", + "name": "Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.59, + "output": 0.79 + } + }, + "llama-3.1-8b-instant": { + "id": "llama-3.1-8b-instant", + "name": "Llama 3.1 8B", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.05, + "output": 0.08 + } + }, + "whisper-large-v3-turbo": { + "id": "whisper-large-v3-turbo", + "name": "Whisper Large V3 Turbo", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", + "modalities": { + "input": ["audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 0, + "output": 0 + } + }, + "whisper-large-v3": { + "id": "whisper-large-v3", + "name": "Whisper", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2023-09-01", + "last_updated": "2025-09-05", + "modalities": { + "input": ["audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 0, + "output": 0 + } + }, + "meta-llama/llama-prompt-guard-2-86m": { + "id": "meta-llama/llama-prompt-guard-2-86m", + "name": "Prompt Guard 2 86M", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-05-29", + "last_updated": "2025-05-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512, + "output": 512 + }, + "status": "beta", + "cost": { + "input": 0.04, + "output": 0.04 + } + }, + "meta-llama/llama-prompt-guard-2-22m": { + "id": "meta-llama/llama-prompt-guard-2-22m", + "name": "Llama Prompt Guard 2 22M", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-05-29", + "last_updated": "2025-05-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512, + "output": 512 + }, + "status": "beta", + "cost": { + "input": 0.03, + "output": 0.03 + } + }, + "meta-llama/llama-4-scout-17b-16e-instruct": { + "id": "meta-llama/llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 8192 + }, + "status": "beta", + "cost": { + "input": 0.11, + "output": 0.34 + } + }, + "openai/gpt-oss-safeguard-20b": { + "id": "openai/gpt-oss-safeguard-20b", + "name": "Safety GPT OSS 20B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-29", + "last_updated": "2026-06-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 65536 + }, + "status": "beta", + "cost": { + "input": 0.075, + "output": 0.3 + } + }, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-10-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 65536 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 65536 + }, + "cost": { + "input": 0.075, + "output": 0.3, + "cache_read": 0.0375 + } + }, + "canopylabs/orpheus-v1-english": { + "id": "canopylabs/orpheus-v1-english", + "name": "Canopy Labs Orpheus V1 English", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "canopylabs", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-12-19", + "last_updated": "2025-12-19", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 4000, + "output": 50000 + }, + "status": "beta" + }, + "canopylabs/orpheus-arabic-saudi": { + "id": "canopylabs/orpheus-arabic-saudi", + "name": "Canopy Labs Orpheus Arabic Saudi", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "canopylabs", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": false, + "limit": { + "context": 4000, + "output": 50000 + }, + "status": "beta" + }, + "groq/compound": { + "id": "groq/compound", + "name": "Compound", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "groq", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-09-04", + "last_updated": "2025-09-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + } + }, + "groq/compound-mini": { + "id": "groq/compound-mini", + "name": "Compound Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "groq", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-09-04", + "last_updated": "2025-09-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + } + }, + "qwen/qwen3-32b": { + "id": "qwen/qwen3-32b", + "name": "Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "default"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2025-06-11", + "last_updated": "2025-06-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 40960 + }, + "status": "beta", + "cost": { + "input": 0.29, + "output": 0.59 + } + } + } + }, + "mixlayer": { + "id": "mixlayer", + "env": ["MIXLAYER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://models.mixlayer.ai/v1", + "name": "Mixlayer", + "doc": "https://docs.mixlayer.com", + "models": { + "qwen/qwen3.5-27b": { + "id": "qwen/qwen3.5-27b", + "name": "Qwen3.5 27B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.3, + "output": 2.4 + } + }, + "qwen/qwen3.5-35b-a3b": { + "id": "qwen/qwen3.5-35b-a3b", + "name": "Qwen3.5 35B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.25, + "output": 1.3 + } + }, + "qwen/qwen3.5-9b": { + "id": "qwen/qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.1, + "output": 0.4 + } + }, + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.6, + "output": 3.6 + } + }, + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen3.5 122B A10B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.4, + "output": 3.2 + } + } + } + }, + "orcarouter": { + "id": "orcarouter", + "env": ["ORCAROUTER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.orcarouter.ai/v1", + "name": "OrcaRouter", + "doc": "https://docs.orcarouter.ai", + "models": { + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + } + }, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 + } + }, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.13, + "output": 0.38 + } + }, + "google/gemini-3.1-pro-preview-customtools": { + "id": "google/gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 4, + "output": 18, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemini-flash-lite-latest": { + "id": "google/gemini-flash-lite-latest", + "name": "Gemini Flash-Lite Latest", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 + } + }, + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 + } + }, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 4, + "output": 18, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.06, + "output": 0.33 + } + }, + "google/gemini-3-pro-preview": { + "id": "google/gemini-3-pro-preview", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 4, + "output": 18, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } + } + }, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 + } + }, + "google/gemini-flash-latest": { + "id": "google/gemini-flash-latest", + "name": "Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.075, + "input_audio": 1 + } + }, + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "z-ai/glm-4.5": { + "id": "z-ai/glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 + } + }, + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "z-ai/glm-4.5-air": { + "id": "z-ai/glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.2, + "output": 1.1, + "cache_read": 0.03, + "cache_write": 0 + } + }, + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.2, + "cache_write": 0 + } + }, + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT-5.2 Pro", + "description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 21, + "output": 168 + } + }, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2021-09-01", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 16385, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 1.5, + "cache_read": 0 + } + }, + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 272000 + }, + "cost": { + "input": 15, + "output": 120 + } + }, + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/gpt-4": { + "id": "openai/gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 30, + "output": 60 + } + }, + "openai/gpt-4o-2024-05-13": { + "id": "openai/gpt-4o-2024-05-13", + "name": "GPT-4o (2024-05-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 5, + "output": 15 + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 + } + }, + "openai/gpt-5-chat-latest": { + "id": "openai/gpt-5-chat-latest", + "name": "GPT-5 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.3-chat-latest": { + "id": "openai/gpt-5.3-chat-latest", + "name": "GPT-5.3 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-4o-2024-08-06": { + "id": "openai/gpt-4o-2024-08-06", + "name": "GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-08-06", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "openai/gpt-5.1-chat-latest": { + "id": "openai/gpt-5.1-chat-latest", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT-5.1 for polished assistants, writing, and product conversations", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.2-chat-latest": { + "id": "openai/gpt-5.2-chat-latest", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "openai/gpt-4o-2024-11-20": { + "id": "openai/gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-11-20", + "last_updated": "2024-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 22.5, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } + } + }, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 + } + }, + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 10, + "output": 30 + } + }, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 + } + }, + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 60, + "output": 270, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } + } + }, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + }, + "kimi/kimi-k2.5": { + "id": "kimi/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.6, + "output": 3, + "cache_read": 0.1 + } + }, + "kimi/kimi-k2.6": { + "id": "kimi/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.16 + } + }, + "orcarouter/auto": { + "id": "orcarouter/auto", + "name": "OrcaRouter Auto", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2026-05-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "grok/grok-4.3": { + "id": "grok/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 30000 + }, + "cost": { + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } + } + }, + "qwen/qwen3.6-35b-a3b": { + "id": "qwen/qwen3.6-35b-a3b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.248, + "output": 1.485 + } + }, + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.359, + "output": 1.434 + } + }, + "qwen/qwen3.5-plus": { + "id": "qwen/qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.115, + "output": 0.688, + "reasoning": 2.4 + } + }, + "qwen/qwen3.5-27b": { + "id": "qwen/qwen3.5-27b", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.086, + "output": 0.688 + } + }, + "qwen/qwen3.5-35b-a3b": { + "id": "qwen/qwen3.5-35b-a3b", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.057, + "output": 0.459 + } + }, + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-15", + "last_updated": "2026-02-15", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.172, + "output": 1.032 + } + }, + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.115, + "output": 0.917 + } + }, + "qwen/qwen3.6-plus": { + "id": "qwen/qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } + } + }, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.19, + "output": 0.37, + "cache_read": 0.0028 + } + }, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.56, + "output": 1.12, + "cache_read": 0.003625 + } + }, + "deepseek/deepseek-reasoner": { + "id": "deepseek/deepseek-reasoner", + "name": "DeepSeek Reasoner", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.028 + } + }, + "deepseek/deepseek-chat": { + "id": "deepseek/deepseek-chat", + "name": "DeepSeek Chat", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 + } + }, + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 + } + }, + "minimax/minimax-m2.5-highspeed": { + "id": "minimax/minimax-m2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-02-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 + } + } + } + }, + "helicone": { + "id": "helicone", + "env": ["HELICONE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://ai-gateway.helicone.ai/v1", + "name": "Helicone", + "doc": "https://helicone.ai/models", + "models": { + "chatgpt-4o-latest": { + "id": "chatgpt-4o-latest", + "name": "OpenAI ChatGPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2024-08-14", + "last_updated": "2024-08-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 5, + "output": 20, + "cache_read": 2.5 + } + }, + "gpt-4.1-mini-2025-04-14": { + "id": "gpt-4.1-mini-2025-04-14", + "name": "OpenAI GPT-4.1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.39999999999999997, + "output": 1.5999999999999999, + "cache_read": 0.09999999999999999 + } + }, + "deepseek-v3.1-terminus": { + "id": "deepseek-v3.1-terminus", + "name": "DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.27, + "output": 1, + "cache_read": 0.21600000000000003 + } + }, + "claude-3.5-haiku": { + "id": "claude-3.5-haiku", + "name": "Anthropic: Claude 3.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "cost": { + "input": 0.7999999999999999, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 + } + }, + "llama-3.1-8b-instruct": { + "id": "llama-3.1-8b-instruct", + "name": "Meta Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 16384, + "output": 16384 + }, + "cost": { + "input": 0.02, + "output": 0.049999999999999996 + } + }, + "o3": { + "id": "o3", + "name": "OpenAI o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "llama-prompt-guard-2-86m": { + "id": "llama-prompt-guard-2-86m", + "name": "Meta Llama Prompt Guard 2 86M", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-01", + "last_updated": "2024-10-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 512, + "output": 2 + }, + "cost": { + "input": 0.01, + "output": 0.01 + } + }, + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.09999999999999999, + "output": 0.3 + } + }, + "hermes-2-pro-llama-3-8b": { + "id": "hermes-2-pro-llama-3-8b", + "name": "Hermes 2 Pro Llama 3 8B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-05", + "release_date": "2024-05-27", + "last_updated": "2024-05-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.14, + "output": 0.14 + } + }, + "deepseek-v3": { + "id": "deepseek-v3", + "name": "DeepSeek V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2024-12-26", + "last_updated": "2024-12-26", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.56, + "output": 1.68, + "cache_read": 0.07 + } + }, + "grok-code-fast-1": { + "id": "grok-code-fast-1", + "name": "xAI Grok Code Fast 1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2024-08-25", + "last_updated": "2024-08-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 10000 + }, + "cost": { + "input": 0.19999999999999998, + "output": 1.5, + "cache_read": 0.02 + } + }, + "o1-mini": { + "id": "o1-mini", + "name": "OpenAI: o1-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 65536 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "deepseek-r1-distill-llama-70b": { + "id": "deepseek-r1-distill-llama-70b", + "name": "DeepSeek R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0.03, + "output": 0.13 + } + }, + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 40960 + }, + "cost": { + "input": 0.29, + "output": 0.59 + } + }, + "claude-sonnet-4": { + "id": "claude-sonnet-4", + "name": "Anthropic: Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-05-14", + "last_updated": "2025-05-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.30000000000000004, + "cache_write": 3.75 + } + }, + "qwen3-next-80b-a3b-instruct": { + "id": "qwen3-next-80b-a3b-instruct", + "name": "Qwen3 Next 80B A3B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262000, + "output": 16384 + }, + "cost": { + "input": 0.14, + "output": 1.4 + } + }, + "llama-4-maverick": { + "id": "llama-4-maverick", + "name": "Meta Llama 4 Maverick 17B 128E", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.15, + "output": 0.6 + } + }, + "mistral-nemo": { + "id": "mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16400 + }, + "cost": { + "input": 20, + "output": 40 + } + }, + "llama-3.3-70b-versatile": { + "id": "llama-3.3-70b-versatile", + "name": "Meta Llama 3.3 70B Versatile", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 32678 + }, + "cost": { + "input": 0.59, + "output": 0.7899999999999999 + } + }, + "gemma2-9b-it": { + "id": "gemma2-9b-it", + "name": "Google Gemma 2", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-06-25", + "last_updated": "2024-06-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0.01, + "output": 0.03 + } + }, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Google Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.3125, + "cache_write": 1.25 + } + }, + "gpt-5": { + "id": "gpt-5", + "name": "OpenAI GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + }, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Anthropic: Claude 4.5 Haiku (20251001)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-10", + "release_date": "2025-10-01", + "last_updated": "2025-10-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.09999999999999999, + "cache_write": 1.25 + } + }, + "claude-4.5-haiku": { + "id": "claude-4.5-haiku", + "name": "Anthropic: Claude 4.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-10", + "release_date": "2025-10-01", + "last_updated": "2025-10-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.09999999999999999, + "cache_write": 1.25 + } + }, + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "OpenAI: GPT-5 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 15, + "output": 120 + } + }, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Google Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65535 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.075, + "cache_write": 0.3 + } + }, + "gpt-4o": { + "id": "gpt-4o", + "name": "OpenAI GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-05", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "o4-mini": { + "id": "o4-mini", + "name": "OpenAI o4 Mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 + } + }, + "llama-3.1-8b-instant": { + "id": "llama-3.1-8b-instant", + "name": "Meta Llama 3.1 8B Instant", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 32678 + }, + "cost": { + "input": 0.049999999999999996, + "output": 0.08 + } + }, + "llama-prompt-guard-2-22m": { + "id": "llama-prompt-guard-2-22m", + "name": "Meta Llama Prompt Guard 2 22M", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-01", + "last_updated": "2024-10-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 512, + "output": 2 + }, + "cost": { + "input": 0.01, + "output": 0.01 + } + }, + "o3-pro": { + "id": "o3-pro", + "name": "OpenAI o3 Pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 20, + "output": 80 + } + }, + "grok-3-mini": { + "id": "grok-3-mini", + "name": "xAI Grok 3 Mini", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 0.5, + "cache_read": 0.075 + } + }, + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "Anthropic: Claude Opus 4.1 (20250805)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "deepseek-tng-r1t2-chimera": { + "id": "deepseek-tng-r1t2-chimera", + "name": "DeepSeek TNG R1T2 Chimera", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-02", + "last_updated": "2025-07-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 130000, + "output": 163840 + }, + "cost": { + "input": 0.3, + "output": 1.2 + } + }, + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Meta Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16400 + }, + "cost": { + "input": 0.13, + "output": 0.39 + } + }, + "sonar-reasoning-pro": { + "id": "sonar-reasoning-pro", + "name": "Perplexity Sonar Reasoning Pro", + "description": "Web-grounded reasoning model for multi-step research and cited answers", + "family": "sonar-reasoning", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 127000, + "output": 4096 + }, + "cost": { + "input": 2, + "output": 8 + } + }, + "grok-3": { + "id": "grok-3", + "name": "xAI Grok 3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.75 + } + }, + "glm-4.6": { + "id": "glm-4.6", + "name": "Zai GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.44999999999999996, + "output": 1.5 + } + }, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 262144 + }, + "cost": { + "input": 0.48, + "output": 2 + } + }, + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "xAI Grok 4.1 Fast Non-Reasoning", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-17", + "last_updated": "2025-11-17", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 30000 + }, + "cost": { + "input": 0.19999999999999998, + "output": 0.5, + "cache_read": 0.049999999999999996 + } + }, + "qwen3-coder": { + "id": "qwen3-coder", + "name": "Qwen3 Coder 480B A35B Instruct Turbo", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 16384 + }, + "cost": { + "input": 0.22, + "output": 0.95 + } + }, + "gpt-5-chat-latest": { + "id": "gpt-5-chat-latest", + "name": "OpenAI GPT-5 Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2024-09", + "release_date": "2024-09-30", + "last_updated": "2024-09-30", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + }, + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "OpenAI: GPT-5.1 Codex", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-codex", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + }, + "o3-mini": { + "id": "o3-mini", + "name": "OpenAI o3 Mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2023-10", + "release_date": "2023-10-01", + "last_updated": "2023-10-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "mistral-large-2411": { + "id": "mistral-large-2411", + "name": "Mistral-Large", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-24", + "last_updated": "2024-07-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 6 + } + }, + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Google Gemini 2.5 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65535 + }, + "cost": { + "input": 0.09999999999999999, + "output": 0.39999999999999997, + "cache_read": 0.024999999999999998, + "cache_write": 0.09999999999999999 + } + }, + "llama-guard-4": { + "id": "llama-guard-4", + "name": "Meta Llama Guard 4 12B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 1024 + }, + "cost": { + "input": 0.21, + "output": 0.21 + } + }, + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Anthropic: Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "claude-3.7-sonnet": { + "id": "claude-3.7-sonnet", + "name": "Anthropic: Claude 3.7 Sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-02", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.30000000000000004, + "cache_write": 3.75 + } + }, + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "OpenAI: GPT-5.1 Codex Mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-codex", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.024999999999999998 + } + }, + "gpt-5.1-chat-latest": { + "id": "gpt-5.1-chat-latest", + "name": "OpenAI GPT-5.1 Chat", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-codex", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + }, + "claude-3-haiku-20240307": { + "id": "claude-3-haiku-20240307", + "name": "Anthropic: Claude 3 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-03", + "release_date": "2024-03-07", + "last_updated": "2024-03-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 4096 + }, + "cost": { + "input": 0.25, + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 + } + }, + "grok-4-fast-reasoning": { + "id": "grok-4-fast-reasoning", + "name": "xAI: Grok 4 Fast Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 2000000 + }, + "cost": { + "input": 0.19999999999999998, + "output": 0.5, + "cache_read": 0.049999999999999996 + } + }, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "OpenAI GPT-4.1 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.09999999999999999, + "output": 0.39999999999999997, + "cache_read": 0.024999999999999998 + } + }, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "OpenAI GPT-OSS 120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.04, + "output": 0.16 + } + }, + "sonar": { + "id": "sonar", + "name": "Perplexity Sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 127000, + "output": 4096 + }, + "cost": { + "input": 1, + "output": 1 + } + }, + "qwen2.5-coder-7b-fast": { + "id": "qwen2.5-coder-7b-fast", + "name": "Qwen2.5 Coder 7B fast", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-09", + "release_date": "2024-09-15", + "last_updated": "2024-09-15", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32000, + "output": 8192 + }, + "cost": { + "input": 0.03, + "output": 0.09 + } + }, + "o1": { + "id": "o1", + "name": "OpenAI: o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 15, + "output": 60, + "cache_read": 7.5 + } + }, + "ernie-4.5-21b-a3b-thinking": { + "id": "ernie-4.5-21b-a3b-thinking", + "name": "Baidu Ernie 4.5 21B A3B Thinking", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "ernie", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-03-16", + "last_updated": "2025-03-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 8000 + }, + "cost": { + "input": 0.07, + "output": 0.28 + } + }, + "llama-4-scout": { + "id": "llama-4-scout", + "name": "Meta Llama 4 Scout 17B 16E", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.08, + "output": 0.3 + } + }, + "sonar-pro": { + "id": "sonar-pro", + "name": "Perplexity Sonar Pro", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "family": "sonar-pro", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 4096 + }, + "cost": { + "input": 3, + "output": 15 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "OpenAI GPT-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Anthropic: Claude Sonnet 4.5 (20250929)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.30000000000000004, + "cache_write": 3.75 + } + }, + "deepseek-reasoner": { + "id": "deepseek-reasoner", + "name": "DeepSeek Reasoner", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 64000 + }, + "cost": { + "input": 0.56, + "output": 1.68, + "cache_read": 0.07 + } + }, + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "xAI Grok 4.1 Fast Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-17", + "last_updated": "2025-11-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 2000000 + }, + "cost": { + "input": 0.19999999999999998, + "output": 0.5, + "cache_read": 0.049999999999999996 + } + }, + "gemini-3-pro-preview": { + "id": "gemini-3-pro-preview", + "name": "Google Gemini 3 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.19999999999999998 + } + }, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "OpenAI GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.024999999999999998 + } + }, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "OpenAI GPT-4.1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.39999999999999997, + "output": 1.5999999999999999, + "cache_read": 0.09999999999999999 + } + }, + "sonar-reasoning": { + "id": "sonar-reasoning", + "name": "Perplexity Sonar Reasoning", + "description": "Web-grounded reasoning model for multi-step research and cited answers", + "family": "sonar-reasoning", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 127000, + "output": 4096 + }, + "cost": { + "input": 1, + "output": 5 + } + }, + "sonar-deep-research": { + "id": "sonar-deep-research", + "name": "Perplexity Sonar Deep Research", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar-deep-research", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 127000, + "output": 4096 + }, + "cost": { + "input": 2, + "output": 8 + } + }, + "kimi-k2-0905": { + "id": "kimi-k2-0905", + "name": "Kimi K2 (09/05)", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 16384 + }, + "cost": { + "input": 0.5, + "output": 2, + "cache_read": 0.39999999999999997 + } + }, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "OpenAI GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.049999999999999996, + "output": 0.39999999999999997, + "cache_read": 0.005 + } + }, + "grok-4": { + "id": "grok-4", + "name": "xAI Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-09", + "last_updated": "2024-07-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.75 + } + }, + "qwen3-235b-a22b-thinking": { + "id": "qwen3-235b-a22b-thinking", + "name": "Qwen3 235B A22B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 81920 + }, + "cost": { + "input": 0.3, + "output": 2.9000000000000004 + } + }, + "claude-opus-4": { + "id": "claude-opus-4", + "name": "Anthropic: Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-05-14", + "last_updated": "2025-05-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "grok-4-fast-non-reasoning": { + "id": "grok-4-fast-non-reasoning", + "name": "xAI Grok 4 Fast Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 2000000 + }, + "cost": { + "input": 0.19999999999999998, + "output": 0.5, + "cache_read": 0.049999999999999996 + } + }, + "claude-4.5-opus": { + "id": "claude-4.5-opus", + "name": "Anthropic: Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "qwen3-vl-235b-a22b-instruct": { + "id": "qwen3-vl-235b-a22b-instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 16384 + }, + "cost": { + "input": 0.3, + "output": 1.5 + } + }, + "kimi-k2-0711": { + "id": "kimi-k2-0711", + "name": "Kimi K2 (07/11)", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 16384 + }, + "cost": { + "input": 0.5700000000000001, + "output": 2.3 + } + }, + "gemma-3-12b-it": { + "id": "gemma-3-12b-it", + "name": "Google Gemma 3 12B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.049999999999999996, + "output": 0.09999999999999999 + } + }, + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "OpenAI GPT-4o-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "OpenAI GPT-OSS 20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.049999999999999996, + "output": 0.19999999999999998 + } + }, + "claude-3.5-sonnet-v2": { + "id": "claude-3.5-sonnet-v2", + "name": "Anthropic: Claude 3.5 Sonnet v2", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 8192 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.30000000000000004, + "cache_write": 3.75 + } + }, + "qwen3-30b-a3b": { + "id": "qwen3-30b-a3b", + "name": "Qwen3 30B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2025-06-01", + "last_updated": "2025-06-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 41000, + "output": 41000 + }, + "cost": { + "input": 0.08, + "output": 0.29 + } + }, + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "OpenAI: GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + }, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 163840, + "output": 65536 + }, + "cost": { + "input": 0.27, + "output": 0.41 + } + }, + "claude-4.5-sonnet": { + "id": "claude-4.5-sonnet", + "name": "Anthropic: Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.30000000000000004, + "cache_write": 3.75 + } + }, + "mistral-small": { + "id": "mistral-small", + "name": "Mistral Small 3.2", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.075, + "output": 0.2 + } + }, + "llama-3.1-8b-instruct-turbo": { + "id": "llama-3.1-8b-instruct-turbo", + "name": "Meta Llama 3.1 8B Instruct Turbo", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 128000 + }, + "cost": { + "input": 0.02, + "output": 0.03 + } + }, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "OpenAI GPT-5.1", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.12500000000000003 + } + } + } + }, + "zai": { + "id": "zai", + "env": ["ZHIPU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.z.ai/api/paas/v4", + "name": "Z.AI", + "doc": "https://docs.z.ai/guides/overview/pricing", + "models": { + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "glm-4.5v": { + "id": "glm-4.5v", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 64000, + "output": 16384 + }, + "cost": { + "input": 0.6, + "output": 1.8 + } + }, + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "glm-4.7-flashx": { + "id": "glm-4.7-flashx", + "name": "GLM-4.7-FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0.07, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0 + } + }, + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 + } + }, + "glm-4.6": { + "id": "glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 + } + }, + "glm-4.6v": { + "id": "glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0.3, + "output": 0.9 + } + }, + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", + "modalities": { + "input": ["text", "image", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.2, + "output": 4, + "cache_read": 0.24, + "cache_write": 0 + } + }, + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.2, + "output": 1.1, + "cache_read": 0.03, + "cache_write": 0 + } + }, + "glm-4.7-flash": { + "id": "glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "glm-4.5-flash": { + "id": "glm-4.5-flash", + "name": "GLM-4.5-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.2, + "cache_write": 0 + } + }, + "glm-5-turbo": { + "id": "glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 131072 + }, + "cost": { + "input": 1.2, + "output": 4, + "cache_read": 0.24, + "cache_write": 0 + } + } + } + }, + "nearai": { + "id": "nearai", + "env": ["NEARAI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://cloud-api.near.ai/v1", + "name": "NEAR AI Cloud", + "doc": "https://docs.near.ai/", + "models": { + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 + } + }, + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.13, + "output": 0.4, + "cache_read": 0.026 + } + }, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + } + }, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 + } + }, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 + } + }, + "google/gemini-3-pro": { + "id": "google/gemini-3-pro", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 15, + "cache_read": 0 + } + }, + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 + } + }, + "Qwen/Qwen3-Embedding-0.6B": { + "id": "Qwen/Qwen3-Embedding-0.6B", + "name": "Qwen3 Embedding 0.6B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-06-03", + "last_updated": "2025-06-03", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 40960, + "output": 1024 + }, + "cost": { + "input": 0.01, + "output": 0 + } + }, + "Qwen/Qwen3-Reranker-0.6B": { + "id": "Qwen/Qwen3-Reranker-0.6B", + "name": "Qwen3 Reranker 0.6B", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-06-03", + "last_updated": "2025-06-03", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 40960, + "output": 1024 + }, + "cost": { + "input": 0.01, + "output": 0.01 + } + }, + "Qwen/Qwen3.6-35B-A3B-FP8": { + "id": "Qwen/Qwen3.6-35B-A3B-FP8", + "name": "Qwen 3.6 35B A3B FP8", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.17, + "output": 1.1, + "cache_read": 0.056 + } + }, + "Qwen/Qwen3.5-122B-A10B": { + "id": "Qwen/Qwen3.5-122B-A10B", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 3.2 + } + }, + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen3 30B-A3B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-29", + "last_updated": "2025-07-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.15, + "output": 0.55 + } + }, + "Qwen/Qwen3-VL-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "name": "Qwen3-VL 30B-A3B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-23", + "last_updated": "2025-09-23", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 32768 + }, + "cost": { + "input": 0.15, + "output": 0.55 + } + }, + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 + } + }, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.8, + "output": 15.5, + "cache_read": 0.18 + } + }, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 + } + }, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT-OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131000, + "output": 32768 + }, + "cost": { + "input": 0.15, + "output": 0.55 + } + }, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } + } + }, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 + } + }, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 + } + }, + "openai/whisper-large-v3": { + "id": "openai/whisper-large-v3", + "name": "Whisper Large v3", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2023-11-06", + "last_updated": "2023-11-06", + "modalities": { + "input": ["audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 448, + "output": 448 + }, + "cost": { + "input": 0.01, + "output": 0 + } + }, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + }, + "anthropic/claude-sonnet-4-5": { + "id": "anthropic/claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15.5, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "zai-org/GLM-5.1-FP8": { + "id": "zai-org/GLM-5.1-FP8", + "name": "GLM-5.1 FP8", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 131072 + }, + "cost": { + "input": 0.85, + "output": 3.3 + } + }, + "black-forest-labs/FLUX.2-klein-4B": { + "id": "black-forest-labs/FLUX.2-klein-4B", + "name": "FLUX.2 Klein 4B", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-01-14", + "last_updated": "2026-01-14", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 128000 + }, + "cost": { + "input": 1, + "output": 1 + } + } + } + }, + "pioneer": { + "id": "pioneer", + "env": ["PIONEER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.pioneer.ai/v1", + "name": "Pioneer", + "doc": "https://agent.pioneer.ai/llms.txt", + "models": { + "gemini-3-flash": { + "id": "gemini-3-flash", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65535 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.083333 + } + }, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.32, + "output": 1.28, + "cache_read": 0.064, + "cache_write": 0.4 + } + }, + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 3.75, + "cache_read": 0.25, + "cache_write": 1.5625 + } + }, + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25, + "cache_write": 2.5 + } + }, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "cache_write": 0.083333 + } + }, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02, + "cache_write": 0.2 + } + }, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.1875, + "output": 1.125, + "cache_read": 0.0375, + "cache_write": 0.234375 + } + }, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "mistral-medium-3.5": { + "id": "mistral-medium-3.5", + "name": "Mistral Medium 3.5", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 1.5, + "output": 7.5, + "cache_read": 1.5, + "cache_write": 1.5 + } + }, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175, + "cache_write": 1.75 + } + }, + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.05, + "cache_write": 0.1 + } + }, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "output": 128000 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 2.5 + } + }, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075, + "cache_write": 0.75 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 1, + "cache_write": 2 + } + }, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025, + "cache_write": 0.25 + } + }, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1047576, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.2, + "cache_write": 0.4 + } + }, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.05, + "output": 0.4, + "cache_read": 0.005, + "cache_write": 0.05 + } + }, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "gemini-3.1-pro": { + "id": "gemini-3.1-pro", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "cache_write": 0.375 + } + }, + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075, + "cache_write": 0.15 + } + }, + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 1.04, + "output": 6.24, + "cache_read": 0.208, + "cache_write": 1.3 + } + }, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 0.325, + "output": 1.95, + "cache_read": 0.065, + "cache_write": 0.40625 + } + }, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "cache_write": 1.25 + } + }, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 5 + } + }, + "LiquidAI/LFM2-24B-A2B": { + "id": "LiquidAI/LFM2-24B-A2B", + "name": "LFM2 24B A2B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "liquid", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-01-31", + "last_updated": "2026-02-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.03, + "output": 0.12, + "cache_read": 0.03, + "cache_write": 0.03 + } + }, + "HuggingFaceTB/SmolLM3-3B-Base": { + "id": "HuggingFaceTB/SmolLM3-3B-Base", + "name": "SmolLM3 3B Base", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "meta-llama/Llama-3.2-1B-Instruct": { + "id": "meta-llama/Llama-3.2-1B-Instruct", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-08-31", + "last_updated": "2024-09-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 60000 + }, + "cost": { + "input": 0.1, + "output": 0.201, + "cache_read": 0.1, + "cache_write": 0.1 + } + }, + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 4096 + }, + "cost": { + "input": 0.9, + "output": 0.9, + "cache_read": 0.9, + "cache_write": 0.9 + } + }, + "meta-llama/Llama-3.1-8B-Instruct": { + "id": "meta-llama/Llama-3.1-8B-Instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-06-30", + "last_updated": "2024-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 16384 + }, + "cost": { + "input": 0.2, + "output": 0.2, + "cache_read": 0.2, + "cache_write": 0.2 + } + }, + "meta-llama/Llama-3.2-3B-Instruct": { + "id": "meta-llama/Llama-3.2-3B-Instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-08-31", + "last_updated": "2024-09-25", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 80000 + }, + "cost": { + "input": 0.1, + "output": 0.335, + "cache_read": 0.1, + "cache_write": 0.1 + } + }, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.34, + "cache_write": 0.95 + } + }, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.95, + "output": 4, + "cache_read": 0.19, + "cache_write": 0.95 + } + }, + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 0.5, + "cache_read": 0.5, + "cache_write": 0.5 + } + }, + "google/diffusiongemma-26B-A4B-it": { + "id": "google/diffusiongemma-26B-A4B-it", + "name": "DiffusionGemma 26B-A4B IT", + "description": "Gemini model for general assistance, reasoning, and multimodal workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-05-31", + "last_updated": "2026-05-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 0.5, + "cache_read": 0.5, + "cache_write": 0.5 + } + }, + "google/gemma-3-4b-pt": { + "id": "google/gemma-3-4b-pt", + "name": "Gemma 3 4B (Pretrained)", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-02-28", + "last_updated": "2025-02-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "google/gemma-4-E4B-it": { + "id": "google/gemma-4-E4B-it", + "name": "Gemma 4 E4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.2, + "output": 0.2, + "cache_read": 0.2, + "cache_write": 0.2 + } + }, + "google/gemma-4-E2B-it": { + "id": "google/gemma-4-E2B-it", + "name": "Gemma 4 E2B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.1, + "output": 0.1, + "cache_read": 0.1, + "cache_write": 0.1 + } + }, + "google/gemma-4-12B-it": { + "id": "google/gemma-4-12B-it", + "name": "Gemma 4 12B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-05-31", + "last_updated": "2026-05-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.25, + "output": 0.25, + "cache_read": 0.25, + "cache_write": 0.25 + } + }, + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0.14, + "output": 1, + "cache_read": 0.028, + "cache_write": 0.175 + } + }, + "Qwen/Qwen3-4B-Instruct-2507": { + "id": "Qwen/Qwen3-4B-Instruct-2507", + "name": "Qwen3 4B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-07-31", + "last_updated": "2025-07-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 4096 + }, + "cost": { + "input": 0.2, + "output": 0.2, + "cache_read": 0.2, + "cache_write": 0.2 + } + }, + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.6, + "output": 0.6, + "cache_read": 0.6, + "cache_write": 0.6 + } + }, + "Qwen/Qwen3-4B-Base": { + "id": "Qwen/Qwen3-4B-Base", + "name": "Qwen3 4B Base", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-03-31", + "last_updated": "2025-03-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "Qwen/Qwen3-1.7B-Base": { + "id": "Qwen/Qwen3-1.7B-Base", + "name": "Qwen3 1.7B Base", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-03-31", + "last_updated": "2025-03-31", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.1, + "output": 0.1, + "cache_read": 0.1, + "cache_write": 0.1 + } + }, + "Qwen/Qwen3-8B": { + "id": "Qwen/Qwen3-8B", + "name": "Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-03-31", + "last_updated": "2025-04-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 4096 + }, + "cost": { + "input": 0.2, + "output": 0.2, + "cache_read": 0.2, + "cache_write": 0.2 + } + }, + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.3, + "output": 0.3, + "cache_read": 0.3, + "cache_write": 0.3 + } + }, + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 16384 + }, + "cost": { + "input": 0.9, + "output": 0.9, + "cache_read": 0.9, + "cache_write": 0.9 + } + }, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.015, + "cache_write": 0.15 + } + }, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.07, + "output": 0.3, + "cache_read": 0.035, + "cache_write": 0.07 + } + }, + "XiaomiMiMo/MiMo-V2.5": { + "id": "XiaomiMiMo/MiMo-V2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "audio", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1050000, + "output": 131072 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028, + "cache_write": 0.14 + } + }, + "XiaomiMiMo/MiMo-V2.5-Pro": { + "id": "XiaomiMiMo/MiMo-V2.5-Pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1050000, + "output": 131072 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036, + "cache_write": 0.435 + } + }, + "mistralai/Mistral-Small-4-119B-2603": { + "id": "mistralai/Mistral-Small-4-119B-2603", + "name": "Mistral Small 4", + "description": "Fast Mistral production model for chat, extraction, and cost-sensitive agents", + "family": "mistral-small", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.015, + "cache_write": 0.15 + } + }, + "mistralai/Mistral-7B-Instruct-v0.3": { + "id": "mistralai/Mistral-7B-Instruct-v0.3", + "name": "Mistral 7B Instruct v0.3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2023-04-30", + "last_updated": "2023-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.2, + "output": 0.2, + "cache_read": 0.2, + "cache_write": 0.2 + } + }, + "mistralai/Mistral-Nemo-Instruct-2407": { + "id": "mistralai/Mistral-Nemo-Instruct-2407", + "name": "Mistral Nemo", + "description": "Efficient Mistral-NVIDIA open model for multilingual chat and local deployment", + "family": "mistral-nemo", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 128000 + }, + "cost": { + "input": 0.02, + "output": 0.03, + "cache_read": 0.02, + "cache_write": 0.02 + } + }, + "pioneer/auto": { + "id": "pioneer/auto", + "name": "Pioneer Auto", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2024-01-01", + "last_updated": "2025-06-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 4096 + } + }, + "sakana/fugu-ultra": { + "id": "sakana/fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "fugu", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "release_date": "2026-06-15", + "last_updated": "2026-06-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 5 + } + }, + "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16": { + "id": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-06-04", + "last_updated": "2026-06-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 65000 + }, + "cost": { + "input": 0.5, + "output": 2.5, + "cache_read": 0.15, + "cache_write": 0.5 + } + }, + "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16": { + "id": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16", + "name": "Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2025-12-15", + "last_updated": "2025-12-15", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0.05, + "output": 0.2, + "cache_read": 0.05, + "cache_write": 0.05 + } + }, + "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8": { + "id": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8", + "name": "Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 32000 + }, + "cost": { + "input": 0.09, + "output": 0.45, + "cache_read": 0.09, + "cache_write": 0.09 + } + }, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1048576, + "output": 128000 + }, + "cost": { + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 1.4 + } + }, + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 202752, + "output": 131072 + }, + "cost": { + "input": 0.98, + "output": 3.08, + "cache_read": 0.182, + "cache_write": 0.98 + } + }, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 0.1, + "output": 0.2, + "cache_read": 0.0197, + "cache_write": 0.1 + } + }, + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625, + "cache_write": 0.435 + } + }, + "fastino/gliner2-multi-large-v1": { + "id": "fastino/gliner2-multi-large-v1", + "name": "GLiNER2 Multi Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-11-30", + "last_updated": "2025-11-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "fastino/gliner2-multi-v1": { + "id": "fastino/gliner2-multi-v1", + "name": "GLiNER2 Multi", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-11-30", + "last_updated": "2025-11-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "fastino/gliguard-LLMGuardrails-300M": { + "id": "fastino/gliguard-LLMGuardrails-300M", + "name": "GLiGuard LLM Guardrails 300M", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "fastino/gliner2-privacy-filter-PII-multi": { + "id": "fastino/gliner2-privacy-filter-PII-multi", + "name": "GLiNER2 Privacy Filter PII (Multi)", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "fastino/gliner2-base-v1": { + "id": "fastino/gliner2-base-v1", + "name": "GLiNER2 Base", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "fastino/gliner2-large-v1": { + "id": "fastino/gliner2-large-v1", + "name": "GLiNER2 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.15, + "cache_read": 0.15, + "cache_write": 0.15 + } + }, + "MiniMaxAI/MiniMax-M3": { + "id": "MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1000000, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.3 + } + }, + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.279, + "output": 1.2, + "cache_read": 0.279, + "cache_write": 0.279 + } + } + } + }, + "llmgateway": { + "id": "llmgateway", + "env": ["LLMGATEWAY_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.llmgateway.io/v1", + "name": "LLM Gateway", + "doc": "https://llmgateway.io/docs", + "models": { + "qwen-coder-plus": { + "id": "qwen-coder-plus", + "name": "Qwen Coder Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2024-09-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.502, + "output": 1.004 + } + }, + "mistral-large-latest": { + "id": "mistral-large-latest", + "name": "Mistral Large (latest)", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 262144 + }, + "cost": { + "input": 4, + "output": 12 + } + }, + "qwen3-vl-235b-a22b-thinking": { + "id": "qwen3-vl-235b-a22b-thinking", + "name": "Qwen3 VL 235B A22B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-09-15", + "last_updated": "2025-09-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.98, + "output": 3.95 + } + }, + "devstral-small-2507": { + "id": "devstral-small-2507", + "name": "Devstral Small", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-07-10", + "last_updated": "2025-07-10", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 128000 + }, + "status": "deprecated", + "cost": { + "input": 0.1, + "output": 0.3 + } + }, + "qwen3-vl-30b-a3b-thinking": { + "id": "qwen3-vl-30b-a3b-thinking", + "name": "Qwen3 VL 30B A3B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-02", + "last_updated": "2025-10-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.2, + "output": 1 + } + }, + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 1050000, + "output": 384000 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 + } + }, + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Hosted Qwen coder for software agents, repo edits, and long-context code", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 6, + "output": 60, + "cache_read": 1.2, + "cache_write": 7.5 + } + }, + "minimax-m2.7-highspeed": { + "id": "minimax-m2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 + } + }, + "qwen-plus": { + "id": "qwen-plus", + "name": "Qwen Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-09-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.4, + "output": 1.2, + "reasoning": 4, + "cache_read": 0.08, + "cache_write": 0.5 + } + }, + "o3": { + "id": "o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "nemotron-3-ultra-550b": { + "id": "nemotron-3-ultra-550b", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-06-04", + "last_updated": "2026-06-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 128000 + }, + "cost": { + "input": 0.5, + "output": 2.5, + "cache_read": 0.15 + } + }, + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 228700, + "output": 131072 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 + } + }, + "grok-4-20-beta-0309-non-reasoning": { + "id": "grok-4-20-beta-0309-non-reasoning", + "name": "Grok 4.20 (Non-Reasoning)", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-09", + "last_updated": "2026-03-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 30000 + }, + "cost": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } + } + }, + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 + }, + "cost": { + "input": 0.38, + "output": 1.98, + "cache_read": 0.19, + "cache_write": 0 + } + }, + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "cache_write": 0.08333 + } + }, + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235B A22B Instruct (2507)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-07-08", + "last_updated": "2025-07-08", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 8192 + }, + "cost": { + "input": 0.09, + "output": 0.58 + } + }, + "llama-3-70b-instruct": { + "id": "llama-3-70b-instruct", + "name": "Llama 3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": true, + "temperature": true, + "release_date": "2024-04-18", + "last_updated": "2024-04-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8000 + }, + "cost": { + "input": 0.51, + "output": 0.74 + } + }, + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262000, + "output": 65536 + }, + "cost": { + "input": 0.07, + "output": 0.27 + } + }, + "seed-1-8-251228": { + "id": "seed-1-8-251228", + "name": "Seed 1.8 (251228)", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-18", + "last_updated": "2025-12-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.05 } - } - } - }, - "moark": { - "id": "moark", - "env": ["MOARK_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://moark.com/v1", - "name": "Moark", - "doc": "https://moark.com/docs/openapi/v1#tag/%E6%96%87%E6%9C%AC%E7%94%9F%E6%88%90", - "models": { - "GLM-4.7": { - "id": "GLM-4.7", - "name": "GLM-4.7", - "family": "glm", + }, + "kimi-k2": { + "id": "kimi-k2", + "name": "Kimi K2", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "structured_output": true, + "temperature": true, + "release_date": "2025-07-11", + "last_updated": "2025-07-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 16384 }, + "cost": { + "input": 0.57, + "output": 2.3, + "cache_read": 0.5 + } + }, + "llama-3.1-70b-instruct": { + "id": "llama-3.1-70b-instruct", + "name": "Llama 3.1 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 2048 }, + "status": "beta", "cost": { - "input": 3.5, - "output": 14 + "input": 0.72, + "output": 0.72 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", + "gpt-5.2-pro": { + "id": "gpt-5.2-pro", + "name": "GPT-5.2 Pro", + "description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 21, + "output": 168 + } + }, + "minimax-m2.1": { + "id": "minimax-m2.1", "name": "MiniMax-M2.1", + "description": "Earlier MiniMax agent model for practical coding and productivity tasks", "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "release_date": "2025-12-23", @@ -27994,269 +67572,1224 @@ "output": 131072 }, "cost": { - "input": 2.1, - "output": 8.4 + "input": 0.27, + "output": 1.1 } - } - } - }, - "opencode-go": { - "id": "opencode-go", - "env": ["OPENCODE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://opencode.ai/zen/go/v1", - "name": "OpenCode Go", - "doc": "https://opencode.ai/docs/zen", - "models": { - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "MiniMax M2.7", - "family": "minimax-m2.7", + }, + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 32768, + "output": 16384 }, - "provider": { - "npm": "@ai-sdk/anthropic" + "cost": { + "input": 0.1, + "output": 0.3, + "reasoning": 8.4 + } + }, + "pixtral-large-latest": { + "id": "pixtral-large-latest", + "name": "Pixtral Large (latest)", + "description": "Mistral's larger vision model for document-heavy image understanding and chat", + "family": "pixtral", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2024-11-04", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 4, + "output": 12 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi-k2.5", + "glm-4-32b-0414-128k": { + "id": "glm-4-32b-0414-128k", + "name": "GLM-4 32B (0414-128k)", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.1, + "output": 0.1 + } + }, + "seed-1-6-flash-250715": { + "id": "seed-1-6-flash-250715", + "name": "Seed 1.6 Flash (250715)", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "structured_output": true, + "temperature": true, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.07, + "output": 0.3, + "cache_read": 0.015 + } + }, + "qwen3-next-80b-a3b-instruct": { + "id": "qwen3-next-80b-a3b-instruct", + "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.15, + "output": 1.2 + } + }, + "qwen3-vl-8b-instruct": { + "id": "qwen3-vl-8b-instruct", + "name": "Qwen3 VL 8B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.08, + "output": 0.5 + } + }, + "gpt-4o-mini-search-preview": { + "id": "gpt-4o-mini-search-preview", + "name": "GPT-4o Mini Search Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6 + } + }, + "glm-4.5v": { + "id": "glm-4.5v", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.6, + "output": 1.8, + "cache_read": 0.11 + } + }, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0.4, + "output": 1.6, + "cache_read": 0.08, + "cache_write": 0.5 + } + }, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + } + }, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 + } + }, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "qwen3.6-35b-a3b": { + "id": "qwen3.6-35b-a3b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, "limit": { "context": 262144, "output": 65536 }, + "cost": { + "input": 0.248, + "output": 1.485 + } + }, + "gpt-3.5-turbo": { + "id": "gpt-3.5-turbo", + "name": "GPT-3.5-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2021-09-01", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 16385, + "output": 4096 + }, + "cost": { + "input": 0.5, + "output": 1.5, + "cache_read": 0 + } + }, + "seed-1-6-250615": { + "id": "seed-1-6-250615", + "name": "Seed 1.6 (250615)", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-06-25", + "last_updated": "2025-06-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.05 + } + }, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 3.75, + "cache_read": 0.125, + "cache_write": 3.125 + } + }, + "glm-4.5": { + "id": "glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131000, + "output": 98304 + }, "cost": { "input": 0.6, - "output": 3, - "cache_read": 0.1 + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 272000 + }, + "cost": { + "input": 15, + "output": 120 + } + }, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.03 + } + }, + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "ministral-8b-2512": { + "id": "ministral-8b-2512", + "name": "Ministral 8B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "mistral", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, + "temperature": true, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 8192 + }, + "cost": { + "input": 0.15, + "output": 0.15 + } + }, + "gpt-4": { + "id": "gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 30, + "output": 60 + } + }, + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 + } + }, + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.845, + "output": 3.38, + "cache_read": 0.6, + "cache_write": 3.75 + } + }, + "mistral-small-2506": { + "id": "mistral-small-2506", + "name": "Mistral Small 3.2", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.1, + "output": 0.3 + } + }, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "cache_write": 0.08333 + } + }, + "qwen-plus-latest": { + "id": "qwen-plus-latest", + "name": "Qwen Plus Latest", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2025-01-25", + "last_updated": "2025-01-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 8192 + }, + "cost": { + "input": 0.4, + "output": 1.2, + "cache_read": 0.08, + "cache_write": 0.5 + } + }, + "muse-spark-1.1": { + "id": "muse-spark-1.1", + "name": "Muse Spark 1.1", + "description": "Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.", + "family": "muse", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-08", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 32000 + }, + "cost": { + "input": 1.25, + "output": 4.25, + "cache_read": 0.15 + } + }, + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 + } + }, + "gemma-4-31b-it": { + "id": "gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.13, + "output": 0.38 + } + }, + "grok-4-3": { + "id": "grok-4-3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 30000 + }, + "cost": { + "input": 1.25, + "output": 2.5, + "cache_read": 0.3125, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } + } + }, + "llama-4-scout-17b-instruct": { + "id": "llama-4-scout-17b-instruct", + "name": "Llama 4 Scout 17B Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-04-05", + "last_updated": "2025-04-05", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 2048 + }, + "cost": { + "input": 0.18, + "output": 0.59 + } + }, + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 4096 + }, + "cost": { + "input": 0.13, + "output": 0.4 + } + }, + "sonar-reasoning-pro": { + "id": "sonar-reasoning-pro", + "name": "Sonar Reasoning Pro", + "description": "Web-grounded Sonar for multi-step research questions that need cited reasoning", + "family": "sonar-reasoning", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 2, + "output": 8 } }, - "mimo-v2.5-pro": { - "id": "mimo-v2.5-pro", - "name": "MiMo V2.5 Pro", - "family": "mimo-v2.5-pro", + "grok-4-20-non-reasoning": { + "id": "grok-4-20-non-reasoning", + "name": "Grok 4.20 (Non-Reasoning)", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 128000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 1, - "output": 3, + "input": 1.25, + "output": 2.5, "cache_read": 0.2, "tiers": [ { - "input": 2, - "output": 6, + "input": 2.5, + "output": 5, "cache_read": 0.4, "tier": { "type": "context", - "size": 256000 + "size": 200000 } } ], "context_over_200k": { - "input": 2, - "output": 6, + "input": 2.5, + "output": 5, "cache_read": 0.4 } } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", + "glm-4.7-flashx": { + "id": "glm-4.7-flashx", + "name": "GLM-4.7-FlashX", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 32768 + "context": 200000, + "output": 131072 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.07, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0 } }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo V2 Omni", - "family": "mimo-v2-omni", - "attachment": true, - "reasoning": true, + "qwen3-30b-a3b-instruct-2507": { + "id": "qwen3-30b-a3b-instruct-2507", + "name": "Qwen3 30B A3B Instruct (2507)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2025-07-08", + "last_updated": "2025-07-08", "modalities": { - "input": ["text", "image", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 128000 + "context": 262000, + "output": 8192 }, - "status": "deprecated", "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 0.1, + "output": 0.3 } }, - "mimo-v2.5": { - "id": "mimo-v2.5", - "name": "MiMo V2.5", - "family": "mimo-v2.5", + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - } - } - }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen3.6", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 262144, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "output": 262144 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, "glm-5.1": { "id": "glm-5.1", "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", "release_date": "2026-04-07", "last_updated": "2026-04-07", "modalities": { @@ -28265,21 +68798,24 @@ }, "open_weights": true, "limit": { - "context": 202752, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.931, + "output": 2.93, + "cache_read": 0.173, + "cache_write": 0 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -28295,523 +68831,549 @@ }, "open_weights": true, "limit": { - "context": 1000000, + "context": 1050000, "output": 384000 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.0028 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", - "attachment": true, + "qwen3-next-80b-a3b-thinking": { + "id": "qwen3-next-80b-a3b-thinking", + "name": "Qwen3-Next 80B-A3B (Thinking)", + "description": "Efficient Qwen thinking model for local reasoning, math, and coding agents", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.15, + "output": 1.2 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "glm-4.6": { + "id": "glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 204800, + "output": 131072 }, + "cost": { + "input": 0.55, + "output": 2.2, + "cache_read": 0.11, + "cache_write": 0 + } + }, + "qwen35-397b-a17b": { + "id": "qwen35-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.0145 + "input": 0.6, + "output": 3.6 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax-m2.5", + "qwen3-235b-a22b-thinking-2507": { + "id": "qwen3-235b-a22b-thinking-2507", + "name": "Qwen3 235B A22B Thinking (2507)", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-07-08", + "last_updated": "2025-07-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 262000, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 0.2, + "output": 0.6 } }, - "mimo-v2-pro": { - "id": "mimo-v2-pro", - "name": "MiMo V2 Pro", - "family": "mimo-v2-pro", - "attachment": true, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 128000 + "context": 262144, + "output": 262144 }, - "status": "deprecated", "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - } + "input": 0.6, + "output": 2.5, + "cache_read": 0.06 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", - "family": "qwen3.5", + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.2, - "output": 1.2, - "cache_read": 0.02, - "cache_write": 0.25 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } - } - } - }, - "databricks": { - "id": "databricks", - "env": ["DATABRICKS_HOST", "DATABRICKS_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://${DATABRICKS_HOST}/ai-gateway/mlflow/v1", - "name": "Databricks", - "doc": "https://docs.databricks.com/aws/en/machine-learning/foundation-models/", - "models": { - "databricks-gpt-oss-120b": { - "id": "databricks-gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + }, + "gemini-pro-latest": { + "id": "gemini-pro-latest", + "name": "Gemini Pro Latest", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-02-27", + "last_updated": "2026-02-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.072, - "output": 0.28 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "databricks-gpt-oss-20b": { - "id": "databricks-gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + "glm-4.5-x": { + "id": "glm-4.5-x", + "name": "GLM-4.5 X", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 16384 }, + "status": "beta", "cost": { - "input": 0.05, - "output": 0.2 + "input": 2.2, + "output": 8.9, + "cache_read": 0.45 } }, - "databricks-claude-haiku-4-5": { - "id": "databricks-claude-haiku-4-5", - "name": "Claude Haiku 4.5 (latest)", - "family": "claude-haiku", + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "databricks-claude-sonnet-4-6": { - "id": "databricks-claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", "family": "claude-sonnet", "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 64000 + "output": 1000000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "databricks-gemini-3-pro": { - "id": "databricks-gemini-3-pro", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "Grok 4.1 Fast Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2025-11-19", + "last_updated": "2025-11-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 2000000 + }, + "cost": { + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "databricks-gemini-3-1-pro": { - "id": "databricks-gemini-3-1-pro", - "name": "Gemini 3.1 Pro Preview Custom Tools", - "family": "gemini-pro", + "grok-4-20-beta-0309-reasoning": { + "id": "grok-4-20-beta-0309-reasoning", + "name": "Grok 4.20 (Reasoning)", + "description": "Reasoning Grok for document-heavy analysis and long-horizon tool use", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 2000000, + "output": 30000 }, "cost": { "input": 2, - "output": 12, + "output": 6, "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - }, "tiers": [ { - "input": 4, - "output": 18, + "input": 2.5, + "output": 5, "cache_read": 0.4, "tier": { "type": "context", "size": 200000 } } - ] + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "databricks-gpt-5-4-mini": { - "id": "databricks-gpt-5-4-mini", - "name": "GPT-5.4 mini", - "family": "gpt-mini", - "attachment": true, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 1.5, - "output": 9, - "cache_read": 0.15 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "context": 1024000, + "output": 131072 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 1.26, + "output": 3.96, + "cache_read": 0.234, + "cache_write": 0 } }, - "databricks-claude-opus-4-6": { - "id": "databricks-claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 30, - "output": 150, - "cache_read": 3, - "cache_write": 37.5 - }, - "provider": { - "body": { - "speed": "fast" - }, - "headers": { - "anthropic-beta": "fast-mode-2026-02-01" - } - } - } - } + "context": 262144, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.15, + "output": 0.9, + "cache_read": 0.015 } }, - "databricks-gemini-2-5-pro": { - "id": "databricks-gemini-2-5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "glm-4.5-airx": { + "id": "glm-4.5-airx", + "name": "GLM-4.5 AirX", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - }, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 1.1, + "output": 4.5, + "cache_read": 0.22 } }, - "databricks-gpt-5-nano": { - "id": "databricks-gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", + "gpt-5-chat-latest": { + "id": "gpt-5-chat-latest", + "name": "GPT-5 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", + "temperature": true, + "knowledge": "2024-09-30", "release_date": "2025-08-07", "last_updated": "2025-08-07", "modalities": { @@ -28825,289 +69387,279 @@ "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "databricks-gpt-5-4": { - "id": "databricks-gpt-5-4", - "name": "GPT-5.4", - "family": "gpt", + "ministral-3b-2512": { + "id": "ministral-3b-2512", + "name": "Ministral 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "mistral", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 131072, + "output": 8192 }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "cost": { + "input": 0.1, + "output": 0.1 + } + }, + "qwen2-5-vl-72b-instruct": { + "id": "qwen2-5-vl-72b-instruct", + "name": "Qwen2.5-VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 32768, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.13, + "output": 0.4 } }, - "databricks-gemini-3-1-flash-lite": { - "id": "databricks-gemini-3-1-flash-lite", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", + "glm-4.6v-flashx": { + "id": "glm-4.6v-flashx", + "name": "GLM-4.6V FlashX", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "output": 16000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 0.04, + "output": 0.4, + "cache_read": 0.004 } }, - "databricks-gpt-5": { - "id": "databricks-gpt-5", - "name": "GPT-5", - "family": "gpt", + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, + "context": 512000, "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 } }, - "databricks-claude-opus-4-1": { - "id": "databricks-claude-opus-4-1", - "name": "Claude Opus 4.1 (latest)", - "family": "claude-opus", - "attachment": true, + "qwen3-vl-plus": { + "id": "qwen3-vl-plus", + "name": "Qwen3-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.2, + "output": 1.6, + "reasoning": 4.8, + "cache_read": 0.04, + "cache_write": 0.25 } }, - "databricks-gpt-5-2": { - "id": "databricks-gpt-5-2", - "name": "GPT-5.2", - "family": "gpt", + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "databricks-gpt-5-5": { - "id": "databricks-gpt-5-5", - "name": "GPT-5.5", - "family": "gpt", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, + "context": 400000, + "input": 272000, "output": 128000 }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 12.5, - "output": 75, - "cache_read": 1.25 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } - }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "databricks-gemini-3-flash": { - "id": "databricks-gemini-3-flash", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", - "attachment": true, + "minimax-m2": { + "id": "minimax-m2", + "name": "MiniMax-M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 196608, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "input_audio": 1 + "input": 0.2, + "output": 1, + "cache_read": 0.03 } }, - "databricks-claude-opus-4-5": { - "id": "databricks-claude-opus-4-5", - "name": "Claude Opus 4.5 (latest)", + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { "input": 5, @@ -29116,176 +69668,169 @@ "cache_write": 6.25 } }, - "databricks-gpt-5-mini": { - "id": "databricks-gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "gpt-5.3-chat-latest": { + "id": "gpt-5.3-chat-latest", + "name": "GPT-5.3 Chat (latest)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "databricks-claude-sonnet-4": { - "id": "databricks-claude-sonnet-4", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", + "grok-4-5": { + "id": "grok-4-5", + "name": "Grok 4.5", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 500000, + "output": 500000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 2, + "output": 6, + "cache_read": 0.5 } }, - "databricks-gpt-5-1": { - "id": "databricks-gpt-5-1", - "name": "GPT-5.1", - "family": "gpt", + "claude-3-opus": { + "id": "claude-3-opus", + "name": "Claude 3 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "structured_output": false, + "temperature": true, + "release_date": "2024-03-04", + "last_updated": "2024-03-04", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "output": 4096 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "databricks-claude-opus-4-7": { - "id": "databricks-claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "qwen-omni-turbo": { + "id": "qwen-omni-turbo", + "name": "Qwen-Omni Turbo", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-01-19", + "last_updated": "2025-03-26", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 30, - "output": 150, - "cache_read": 3, - "cache_write": 37.5 - }, - "provider": { - "body": { - "speed": "fast" - }, - "headers": { - "anthropic-beta": "fast-mode-2026-02-01" - } - } - } - } + "context": 32768, + "output": 2048 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.2, + "output": 0.8 } }, - "databricks-gemini-2-5-flash": { - "id": "databricks-gemini-2-5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "input_audio": 1 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "databricks-gpt-5-4-nano": { - "id": "databricks-gpt-5-4-nano", - "name": "GPT-5.4 nano", - "family": "gpt-nano", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -29297,1068 +69842,1356 @@ "output": 128000 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "databricks-claude-sonnet-4-5": { - "id": "databricks-claude-sonnet-4-5", - "name": "Claude Sonnet 4.5 (latest)", - "family": "claude-sonnet", + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } - } - } - }, - "io-net": { - "id": "io-net", - "env": ["IOINTELLIGENCE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.intelligence.io.solutions/api/v1", - "name": "IO.NET", - "doc": "https://io.net/docs/guides/intelligence/io-intelligence", - "models": { - "Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar": { - "id": "Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar", - "name": "Qwen 3 Coder 480B", - "family": "qwen", + }, + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-15", - "last_updated": "2025-01-15", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 106000, - "output": 4096 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.22, - "output": 0.95, - "cache_read": 0.11, - "cache_write": 0.44 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "Qwen/Qwen3-Next-80B-A3B-Instruct": { - "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "name": "Qwen 3 Next 80B Instruct", + "qwen-flash": { + "id": "qwen-flash", + "name": "Qwen Flash", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-10", - "last_updated": "2025-01-10", + "knowledge": "2024-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 4096 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.8, - "cache_read": 0.05, - "cache_write": 0.2 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0.0625 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen 3 235B Thinking", + "qwen3-4b-fp8": { + "id": "qwen3-4b-fp8", + "name": "Qwen3 4B FP8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 4096 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.11, - "output": 0.6, - "cache_read": 0.055, - "cache_write": 0.22 + "input": 0.03, + "output": 0.03 } }, - "Qwen/Qwen2.5-VL-32B-Instruct": { - "id": "Qwen/Qwen2.5-VL-32B-Instruct", - "name": "Qwen 2.5 VL 32B Instruct", + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", + "modalities": { + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.1, + "output": 0.4, + "cache_read": 0.01 + } + }, + "qwen2-5-vl-32b-instruct": { + "id": "qwen2-5-vl-32b-instruct", + "name": "Qwen2.5 VL 32B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "release_date": "2025-03-15", + "last_updated": "2025-03-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.22, - "cache_read": 0.025, - "cache_write": 0.1 + "input": 1.4, + "output": 4.2 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "GLM 4.6", - "family": "glm", - "attachment": false, - "reasoning": false, + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-11-15", - "last_updated": "2024-11-15", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 1.75, - "cache_read": 0.2, - "cache_write": 0.8 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "mistralai/Magistral-Small-2506": { - "id": "mistralai/Magistral-Small-2506", - "name": "Magistral Small 2506", - "family": "magistral-small", - "attachment": false, - "reasoning": false, + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 1.5, - "cache_read": 0.25, - "cache_write": 1 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "mistralai/Mistral-Large-Instruct-2411": { - "id": "mistralai/Mistral-Large-Instruct-2411", - "name": "Mistral Large Instruct 2411", - "family": "mistral-large", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 1, - "cache_write": 4 + "input": 0.405, + "output": 1.98, + "cache_read": 0.225 } }, - "mistralai/Mistral-Nemo-Instruct-2407": { - "id": "mistralai/Mistral-Nemo-Instruct-2407", - "name": "Mistral Nemo Instruct 2407", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, + "seed-1-6-250915": { + "id": "seed-1-6-250915", + "name": "Seed 1.6 (250915)", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-05", - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 256000, + "output": 8192 }, "cost": { - "input": 0.02, - "output": 0.04, - "cache_read": 0.01, - "cache_write": 0.04 + "input": 0.25, + "output": 2, + "cache_read": 0.05 } }, - "mistralai/Devstral-Small-2505": { - "id": "mistralai/Devstral-Small-2505", - "name": "Devstral Small 2505", - "family": "devstral", - "attachment": false, - "reasoning": false, + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-05-01", - "last_updated": "2025-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.22, - "cache_read": 0.025, - "cache_write": 0.1 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "gpt-5.2-chat-latest": { + "id": "gpt-5.2-chat-latest", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 16384 }, "cost": { - "input": 0.13, - "output": 0.38, - "cache_read": 0.065, - "cache_write": 0.26 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { - "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", - "name": "Llama 4 Maverick 17B 128E Instruct", - "family": "llama", + "minimax-text-01": { + "id": "minimax-text-01", + "name": "MiniMax Text 01", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-12", "release_date": "2025-01-15", "last_updated": "2025-01-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 430000, - "output": 4096 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.075, - "cache_write": 0.3 + "input": 0.2, + "output": 1.1 } }, - "meta-llama/Llama-3.2-90B-Vision-Instruct": { - "id": "meta-llama/Llama-3.2-90B-Vision-Instruct", - "name": "Llama 3.2 90B Vision Instruct", - "family": "llama", - "attachment": false, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16000, - "output": 4096 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.35, + "input": 0.1, "output": 0.4, - "cache_read": 0.175, - "cache_write": 0.7 + "cache_read": 0.025 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek R1", - "family": "deepseek-thinking", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-28", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32766 }, "cost": { - "input": 2, - "output": 8.75, - "cache_read": 1, - "cache_write": 4 + "input": 0.05, + "output": 0.25 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT-OSS 20B", - "family": "gpt-oss", + "qwen-vl-max": { + "id": "qwen-vl-max", + "name": "Qwen-VL Max", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "knowledge": "2024-04", + "release_date": "2024-04-08", + "last_updated": "2025-08-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.03, - "output": 0.14, - "cache_read": 0.015, - "cache_write": 0.06 + "input": 0.8, + "output": 3.2 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT-OSS 120B", - "family": "gpt-oss", + "llama-3-8b-instruct": { + "id": "llama-3-8b-instruct", + "name": "Llama 3 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 4096 + "context": 8192, + "output": 8192 }, "cost": { "input": 0.04, - "output": 0.4, - "cache_read": 0.02, - "cache_write": 0.08 + "output": 0.04 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.55, - "output": 2.25, - "cache_read": 0.275, - "cache_write": 1.1 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi K2 Instruct", - "family": "kimi", + "sonar": { + "id": "sonar", + "name": "Sonar", + "description": "Fast web-grounded Sonar for current answers, citations, and lightweight retrieval", + "family": "sonar", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-09-05", - "last_updated": "2024-09-05", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, + "context": 130000, "output": 4096 }, "cost": { - "input": 0.39, - "output": 1.9, - "cache_read": 0.195, - "cache_write": 0.78 + "input": 1, + "output": 1 } - } - } - }, - "alibaba-cn": { - "id": "alibaba-cn", - "env": ["DASHSCOPE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://dashscope.aliyuncs.com/compatible-mode/v1", - "name": "Alibaba (China)", - "doc": "https://www.alibabacloud.com/help/en/model-studio/models", - "models": { - "qwen3-235b-a22b": { - "id": "qwen3-235b-a22b", - "name": "Qwen3 235B-A22B", + }, + "qwen-max": { + "id": "qwen-max", + "name": "Qwen Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-04", + "release_date": "2024-04-03", + "last_updated": "2025-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.287, - "output": 1.147, - "reasoning": 2.868 + "input": 1.6, + "output": 6.4 } }, - "qwen-plus-character": { - "id": "qwen-plus-character", - "name": "Qwen Plus Character", - "family": "qwen", - "attachment": false, - "reasoning": false, + "o1": { + "id": "o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01", - "last_updated": "2024-01", + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 4096 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.115, - "output": 0.287 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "qwen2-5-math-7b-instruct": { - "id": "qwen2-5-math-7b-instruct", - "name": "Qwen2.5-Math 7B Instruct", - "family": "qwen", + "minimax-m2.5-highspeed": { + "id": "minimax-m2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2026-02-13", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 4096, - "output": 3072 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.144, - "output": 0.287 + "input": 0.6, + "output": 2.4, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Moonshot Kimi K2.5", - "family": "kimi", - "attachment": false, + "deepseek-v3.1": { + "id": "deepseek-v3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 128000, "output": 32768 }, "cost": { - "input": 0.574, - "output": 2.411 + "input": 0.56, + "output": 1.68, + "cache_read": 0.112 } }, - "qwen-doc-turbo": { - "id": "qwen-doc-turbo", - "name": "Qwen Doc Turbo", - "family": "qwen", - "attachment": false, + "ministral-14b-2512": { + "id": "ministral-14b-2512", + "name": "Ministral 14B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "mistral", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01", - "last_updated": "2024-01", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, + "context": 262144, "output": 8192 }, "cost": { - "input": 0.087, - "output": 0.144 + "input": 0.2, + "output": 0.2 } }, - "qwen-vl-ocr": { - "id": "qwen-vl-ocr", - "name": "Qwen-VL OCR", - "family": "qwen", - "attachment": false, + "sonar-pro": { + "id": "sonar-pro", + "name": "Sonar Pro", + "description": "Deeper Sonar search model with broader retrieval and stronger synthesis", + "family": "sonar-pro", + "attachment": true, "reasoning": false, "tool_call": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-10-28", - "last_updated": "2025-04-13", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 34096, - "output": 4096 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0.717, - "output": 0.717 + "input": 3, + "output": 15 } }, - "qwen-omni-turbo-realtime": { - "id": "qwen-omni-turbo-realtime", - "name": "Qwen-Omni Turbo Realtime", - "family": "qwen", - "attachment": false, + "glm-4.6v": { + "id": "glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.3, + "output": 0.9, + "cache_read": 0.05 + } + }, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "claude-haiku-4-5-free": { + "id": "claude-haiku-4-5-free", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-05-08", - "last_updated": "2025-05-08", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "audio"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 2048 + "context": 200000, + "output": 200000 }, "cost": { - "input": 0.23, - "output": 0.918, - "input_audio": 3.584, - "output_audio": 7.168 + "input": 0, + "output": 0 } }, - "qwen3-8b": { - "id": "qwen3-8b", - "name": "Qwen3 8B", - "family": "qwen", + "minimax-m2.1-lightning": { + "id": "minimax-m2.1-lightning", + "name": "MiniMax M2.1 Lightning", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 196608, + "output": 131072 }, "cost": { - "input": 0.072, - "output": 0.287, - "reasoning": 0.717 + "input": 0.12, + "output": 0.48 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen3.5 397B-A17B", - "family": "qwen", - "attachment": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } + }, + "mimo-v2.5": { + "id": "mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.43, - "output": 2.58, - "reasoning": 2.58 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028, + "tiers": [ + { + "input": 0.8, + "output": 4, + "cache_read": 0.16, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.8, + "output": 4, + "cache_read": 0.16 + } } }, - "qwen-math-turbo": { - "id": "qwen-math-turbo", - "name": "Qwen Math Turbo", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 3072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "qwq-plus": { - "id": "qwq-plus", - "name": "QwQ Plus", - "family": "qwen", - "attachment": false, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.23, - "output": 0.574 + "input": 0.4, + "output": 2.2, + "cache_read": 0.08 } }, - "qwen-vl-plus": { - "id": "qwen-vl-plus", - "name": "Qwen-VL Plus", - "family": "qwen", - "attachment": false, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-08-15", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.115, - "output": 0.287 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.86, - "output": 3.15 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "deepseek-r1-distill-llama-70b": { - "id": "deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", - "attachment": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.287, - "output": 1.147, - "reasoning": 2.868 + "input": 0.108, + "output": 0.675, + "cache_read": 0.06 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "llama-4-maverick-17b-instruct": { + "id": "llama-4-maverick-17b-instruct", + "name": "Llama 4 Maverick 17B Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 1048576, + "output": 2048 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.27, + "output": 0.85 } }, - "qwen-max": { - "id": "qwen-max", - "name": "Qwen Max", + "qwen3-coder-480b-a35b-instruct": { + "id": "qwen3-coder-480b-a35b-instruct", + "name": "Qwen3-Coder 480B-A35B Instruct", + "description": "Open Qwen coding heavyweight for repository reasoning and agentic engineering", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-03", - "last_updated": "2025-01-25", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.345, - "output": 1.377 + "input": 0.3, + "output": 1.3 } }, - "qwen-plus": { - "id": "qwen-plus", - "name": "Qwen Plus", - "family": "qwen", + "devstral-2512": { + "id": "devstral-2512", + "name": "Devstral 2", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-09-11", + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 32768 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.115, - "output": 0.287, - "reasoning": 1.147 + "input": 0.4, + "output": 2 } }, - "qwen-omni-turbo": { - "id": "qwen-omni-turbo", - "name": "Qwen-Omni Turbo", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01-19", - "last_updated": "2025-03-26", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 2048 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.058, - "output": 0.23, - "input_audio": 3.584, - "output_audio": 7.168 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "qwen-flash": { - "id": "qwen-flash", - "name": "Qwen Flash", - "family": "qwen", - "attachment": false, + "gemma-4-26b-a4b-it": { + "id": "gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, + "context": 262144, "output": 32768 }, "cost": { - "input": 0.022, - "output": 0.216 + "input": 0.07, + "output": 0.34 } }, - "qwen2-5-vl-7b-instruct": { - "id": "qwen2-5-vl-7b-instruct", - "name": "Qwen2.5-VL 7B Instruct", + "qwen3.5-9b": { + "id": "qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.287, - "output": 0.717 + "input": 0.1, + "output": 0.15 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek R1", - "family": "deepseek-thinking", + "qwq-plus": { + "id": "qwq-plus", + "name": "QwQ Plus", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-04", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { "input": ["text"], "output": ["text"] @@ -30366,527 +71199,596 @@ "open_weights": false, "limit": { "context": 131072, - "output": 16384 + "output": 8192 }, "cost": { - "input": 0.574, - "output": 2.294 + "input": 0.8, + "output": 2.4 } }, - "qwen3.5-flash": { - "id": "qwen3.5-flash", - "name": "Qwen3.5 Flash", - "family": "qwen", + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "Grok 4.1 Fast Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-23", - "last_updated": "2026-02-23", + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.172, - "output": 1.72, - "reasoning": 1.72 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", + "qwen3-vl-30b-a3b-instruct": { + "id": "qwen3-vl-30b-a3b-instruct", + "name": "Qwen3 VL 30B A3B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, - "reasoning": true, + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2025-10-02", + "last_updated": "2025-10-02", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 65536 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5 - } + "input": 0.2, + "output": 0.7 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3 Max", + "qwen3-vl-flash": { + "id": "qwen3-vl-flash", + "name": "Qwen3 VL Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "release_date": "2025-10-09", + "last_updated": "2025-10-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 262144, - "output": 65536 + "output": 32000 }, "cost": { - "input": 0.861, - "output": 3.441 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "release_date": "2026-04-14", - "last_updated": "2026-04-14", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 202752, - "output": 128000 - }, - "cost": { - "input": 0.87, - "output": 3.48, - "cache_read": 0.17 - } - }, - "qwen3-omni-flash": { - "id": "qwen3-omni-flash", - "name": "Qwen3-Omni Flash", - "family": "qwen", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] - }, "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.058, - "output": 0.23, - "input_audio": 3.584, - "output_audio": 7.168 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "deepseek-v3-1": { - "id": "deepseek-v3-1", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": false, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.574, - "output": 1.721 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "qwen2-5-72b-instruct": { - "id": "qwen2-5-72b-instruct", - "name": "Qwen2.5 72B Instruct", - "family": "qwen", + "llama-3.1-nemotron-ultra-253b": { + "id": "llama-3.1-nemotron-ultra-253b", + "name": "Llama 3.1 Nemotron Ultra 253B", + "description": "Flagship Nemotron model for high-throughput reasoning and complex agents", + "family": "nemotron", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2025-04-07", + "last_updated": "2025-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 128000, "output": 8192 }, "cost": { - "input": 0.574, - "output": 1.721 + "input": 0.6, + "output": 1.8 } }, - "qwen3-vl-235b-a22b": { - "id": "qwen3-vl-235b-a22b", - "name": "Qwen3-VL 235B-A22B", - "family": "qwen", - "attachment": false, - "reasoning": true, + "gpt-4-turbo": { + "id": "gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.286705, - "output": 1.14682, - "reasoning": 2.867051 + "input": 10, + "output": 30 } }, - "qwen-math-plus": { - "id": "qwen-math-plus", - "name": "Qwen Math Plus", + "qwen3-coder-flash": { + "id": "qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-08-16", - "last_updated": "2024-09-19", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 3072 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.574, - "output": 1.721 + "input": 0.3, + "output": 1.5, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "qwen2-5-coder-32b-instruct": { - "id": "qwen2-5-coder-32b-instruct", - "name": "Qwen2.5-Coder 32B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "grok-build-0-1": { + "id": "grok-build-0-1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-11", - "last_updated": "2024-11", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 1, + "output": 2, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 4, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 4, + "cache_read": 0.4 + } } }, - "qwen3-asr-flash": { - "id": "qwen3-asr-flash", - "name": "Qwen3-ASR Flash", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-09-08", - "last_updated": "2025-09-08", + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 53248, - "output": 4096 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.032, - "output": 0.032 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "qwen-deep-research": { - "id": "qwen-deep-research", - "name": "Qwen Deep Research", - "family": "qwen", - "attachment": false, + "grok-4": { + "id": "grok-4", + "name": "Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01", - "last_updated": "2024-01", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 256000, + "output": 256000 }, "cost": { - "input": 7.742, - "output": 23.367 + "input": 3, + "output": 15, + "cache_read": 0.75 } }, - "qwen3-next-80b-a3b-thinking": { - "id": "qwen3-next-80b-a3b-thinking", - "name": "Qwen3-Next 80B-A3B (Thinking)", - "family": "qwen", + "llama-3.2-11b-instruct": { + "id": "llama-3.2-11b-instruct", + "name": "Llama 3.2 11B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09", - "last_updated": "2025-09", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.144, - "output": 1.434 + "input": 0.07, + "output": 0.33 } }, - "qwen-mt-plus": { - "id": "qwen-mt-plus", - "name": "Qwen-MT Plus", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01", - "last_updated": "2025-01", + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 8192 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.259, - "output": 0.775 + "input": 30, + "output": 180 } }, - "deepseek-r1-distill-qwen-32b": { - "id": "deepseek-r1-distill-qwen-32b", - "name": "DeepSeek R1 Distill Qwen 32B", - "family": "qwen", + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "output": 16384 + "context": 131000, + "output": 98304 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 0.13, + "output": 0.85, + "cache_read": 0.025, + "cache_write": 0 } }, - "qwen-vl-max": { - "id": "qwen-vl-max", - "name": "Qwen-VL Max", - "family": "qwen", - "attachment": false, - "reasoning": false, + "kimi-k2.7-code-highspeed": { + "id": "kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code Highspeed", + "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-08", - "last_updated": "2025-08-13", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.23, - "output": 0.574 + "input": 1.9, + "output": 8, + "cache_read": 0.38 } }, - "qwen3-coder-flash": { - "id": "qwen3-coder-flash", - "name": "Qwen3 Coder Flash", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 65536 + "output": 64000 }, "cost": { - "input": 0.144, - "output": 0.574 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "deepseek-r1-distill-qwen-7b": { - "id": "deepseek-r1-distill-qwen-7b", - "name": "DeepSeek R1 Distill Qwen 7B", - "family": "qwen", - "attachment": false, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.072, - "output": 0.144 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "qwen2-5-7b-instruct": { - "id": "qwen2-5-7b-instruct", - "name": "Qwen2.5 7B Instruct", - "family": "qwen", + "glm-4.7-flash": { + "id": "glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.072, - "output": 0.144 + "input": 0.06, + "output": 0.4, + "cache_read": 0.01, + "cache_write": 0 } }, - "qwen2-5-14b-instruct": { - "id": "qwen2-5-14b-instruct", - "name": "Qwen2.5 14B Instruct", + "qwen3-vl-235b-a22b-instruct": { + "id": "qwen3-vl-235b-a22b-instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, @@ -30895,1207 +71797,1418 @@ "output": 8192 }, "cost": { - "input": 0.144, - "output": 0.431 + "input": 0.3, + "output": 1.5 } }, - "tongyi-intent-detect-v3": { - "id": "tongyi-intent-detect-v3", - "name": "Tongyi Intent Detect V3", - "family": "yi", - "attachment": false, + "gpt-4o-search-preview": { + "id": "gpt-4o-search-preview", + "name": "GPT-4o Search Preview", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01", - "last_updated": "2024-01", + "release_date": "2024-10-01", + "last_updated": "2024-10-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1024 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.058, - "output": 0.144 + "input": 2.5, + "output": 10 } }, - "qwq-32b": { - "id": "qwq-32b", - "name": "QwQ 32B", - "family": "qwen", + "llama-3.2-3b-instruct": { + "id": "llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-12", - "last_updated": "2024-12", + "release_date": "2024-09-18", + "last_updated": "2024-09-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 32768, + "output": 32000 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 0.03, + "output": 0.05 } }, - "moonshot-kimi-k2-instruct": { - "id": "moonshot-kimi-k2-instruct", - "name": "Moonshot Kimi K2 Instruct", - "family": "kimi", + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.574, - "output": 2.294 + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } } }, - "qwen2-5-32b-instruct": { - "id": "qwen2-5-32b-instruct", - "name": "Qwen2.5 32B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.5-pro": { + "id": "gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.287, - "output": 0.861 + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } } }, - "qwen3-next-80b-a3b-instruct": { - "id": "qwen3-next-80b-a3b-instruct", - "name": "Qwen3-Next 80B-A3B Instruct", - "family": "qwen", - "attachment": false, + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09", - "last_updated": "2025-09", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.144, - "output": 0.574 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "qwen3-omni-flash-realtime": { - "id": "qwen3-omni-flash-realtime", - "name": "Qwen3-Omni Flash Realtime", - "family": "qwen", + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 131072, + "output": 32766 }, "cost": { - "input": 0.23, - "output": 0.918, - "input_audio": 3.584, - "output_audio": 7.168 + "input": 0.04, + "output": 0.15 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Moonshot Kimi K2.6", - "family": "kimi", - "attachment": true, + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 203000, + "output": 131072 }, "cost": { - "input": 0.929, - "output": 3.858 + "input": 0.72, + "output": 2.3, + "cache_read": 0.144, + "cache_write": 0 } }, - "qwen3-vl-30b-a3b": { - "id": "qwen3-vl-30b-a3b", - "name": "Qwen3-VL 30B-A3B", + "qwen-max-latest": { + "id": "qwen-max-latest", + "name": "Qwen Max Latest", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, - "reasoning": true, + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-01-25", + "last_updated": "2025-01-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.108, - "output": 0.431, - "reasoning": 1.076 + "input": 1.6, + "output": 6.4 } }, - "qwen3-vl-plus": { - "id": "qwen3-vl-plus", - "name": "Qwen3-VL Plus", - "family": "qwen", - "attachment": false, - "reasoning": true, + "mistral-large-2512": { + "id": "mistral-large-2512", + "name": "Mistral Large 3", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 262144, - "output": 32768 + "output": 262144 }, "cost": { - "input": 0.143353, - "output": 1.433525, - "reasoning": 4.300576 + "input": 0.5, + "output": 1.5 } }, - "deepseek-v3-2-exp": { - "id": "deepseek-v3-2-exp", - "name": "DeepSeek V3.2 Exp", - "family": "deepseek", + "qwen-turbo": { + "id": "qwen-turbo", + "name": "Qwen Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-04", + "release_date": "2024-11-01", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0.287, - "output": 0.431 + "input": 0.05, + "output": 0.2, + "reasoning": 0.5 } }, - "qwen3-coder-480b-a35b-instruct": { - "id": "qwen3-coder-480b-a35b-instruct", - "name": "Qwen3-Coder 480B-A35B Instruct", - "family": "qwen", - "attachment": false, + "custom": { + "id": "custom", + "name": "Custom Model", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.861, - "output": 3.441 + "input": 0, + "output": 0 } }, - "deepseek-r1-distill-qwen-1-5b": { - "id": "deepseek-r1-distill-qwen-1-5b", - "name": "DeepSeek R1 Distill Qwen 1.5B", - "family": "qwen", - "attachment": false, + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3-Coder 30B-A3B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "grok-4-20-reasoning": { + "id": "grok-4-20-reasoning", + "name": "Grok 4.20 (Reasoning)", + "description": "Reasoning Grok for document-heavy analysis and long-horizon tool use", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2026-03-09", + "last_updated": "2026-03-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.216, - "output": 0.861 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "qwen2-5-coder-7b-instruct": { - "id": "qwen2-5-coder-7b-instruct", - "name": "Qwen2.5-Coder 7B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-11", - "last_updated": "2024-11", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 163840, + "output": 16384 }, "cost": { - "input": 0.144, - "output": 0.287 + "input": 0.26, + "output": 0.38, + "cache_read": 0.13 } }, - "qwen-turbo": { - "id": "qwen-turbo", - "name": "Qwen Turbo", + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-11-01", - "last_updated": "2025-07-15", + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.044, - "output": 0.087, - "reasoning": 0.431 + "input": 1.3, + "output": 7.8, + "cache_read": 0.13, + "cache_write": 1.625 } }, - "qwen-mt-turbo": { - "id": "qwen-mt-turbo", - "name": "Qwen-MT Turbo", - "family": "qwen", - "attachment": false, + "auto": { + "id": "auto", + "name": "Auto Route", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01", - "last_updated": "2025-01", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.101, - "output": 0.28 + "input": 0, + "output": 0 } }, - "qwen2-5-math-72b-instruct": { - "id": "qwen2-5-math-72b-instruct", - "name": "Qwen2.5-Math 72B Instruct", - "family": "qwen", + "codestral-2508": { + "id": "codestral-2508", + "name": "Codestral", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "mistral", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 4096, - "output": 3072 + "context": 256000, + "output": 16384 }, "cost": { - "input": 0.574, - "output": 1.721 + "input": 0.3, + "output": 0.9 } }, - "qwen3.6-max-preview": { - "id": "qwen3.6-max-preview", - "name": "Qwen3.6 Max Preview", + "qwen3-235b-a22b-fp8": { + "id": "qwen3-235b-a22b-fp8", + "name": "Qwen3 235B A22B FP8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-21", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, - "limit": { - "context": 245800, - "output": 65536 - }, - "cost": { - "input": 1.32, - "output": 7.9, - "cache_read": 0.132 - } - }, - "qwen2-5-omni-7b": { - "id": "qwen2-5-omni-7b", - "name": "Qwen2.5-Omni 7B", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-12", - "last_updated": "2024-12", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] - }, "open_weights": true, "limit": { - "context": 32768, - "output": 2048 + "context": 40960, + "output": 8192 }, "cost": { - "input": 0.087, - "output": 0.345, - "input_audio": 5.448 + "input": 0.2, + "output": 0.8 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", - "family": "qwen", - "attachment": false, + "fugu-ultra": { + "id": "fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "release_date": "2026-06-22", + "last_updated": "2026-06-22", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 65536 + "output": 1000000 }, "cost": { - "input": 0.573, - "output": 3.44, - "reasoning": 3.44 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "deepseek-r1-distill-qwen-14b": { - "id": "deepseek-r1-distill-qwen-14b", - "name": "DeepSeek R1 Distill Qwen 14B", - "family": "qwen", - "attachment": false, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.144, - "output": 0.431 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "qwen2-5-vl-72b-instruct": { - "id": "qwen2-5-vl-72b-instruct", - "name": "Qwen2.5-VL 72B Instruct", + "qwen-vl-plus": { + "id": "qwen-vl-plus", + "name": "Qwen-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "release_date": "2024-01-25", + "last_updated": "2025-08-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, "output": 8192 }, "cost": { - "input": 2.294, - "output": 6.881 + "input": 0.21, + "output": 0.64 } }, - "deepseek-v3": { - "id": "deepseek-v3", - "name": "DeepSeek V3", - "family": "deepseek", + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.287, - "output": 1.147 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek-thinking", - "attachment": false, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.574, - "output": 2.294 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "qvq-max": { - "id": "qvq-max", - "name": "QVQ Max", - "family": "qvq", - "attachment": false, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 1.147, - "output": 4.588 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } - }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Moonshot Kimi K2 Thinking", - "family": "kimi", + } + } + }, + "alibaba-coding-plan-cn": { + "id": "alibaba-coding-plan-cn", + "env": ["ALIBABA_CODING_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://coding.dashscope.aliyuncs.com/v1", + "name": "Alibaba Coding Plan (China)", + "doc": "https://help.aliyun.com/zh/model-studio/coding-plan", + "models": { + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.574, - "output": 2.294 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "qwen3-14b": { - "id": "qwen3-14b", - "name": "Qwen3 14B", - "family": "qwen", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.144, - "output": 0.574, - "reasoning": 1.434 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-r1-distill-llama-8b": { - "id": "deepseek-r1-distill-llama-8b", - "name": "DeepSeek R1 Distill Llama 8B", - "family": "deepseek-thinking", + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "qwen-long": { - "id": "qwen-long", - "name": "Qwen Long", + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 10000000, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.072, - "output": 0.287 + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 } }, - "kimi/kimi-k2.5": { - "id": "kimi/kimi-k2.5", - "name": "kimi/kimi-k2.5", - "family": "kimi", - "attachment": false, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 262144 - }, - "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 - } - }, - "MiniMax/MiniMax-M2.7": { - "id": "MiniMax/MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.1875, + "output": 1.125, + "cache_write": 0.234375 } }, - "siliconflow/deepseek-v3-0324": { - "id": "siliconflow/deepseek-v3-0324", - "name": "siliconflow/deepseek-v3-0324", - "family": "deepseek", + "qwen3-max-2026-01-23": { + "id": "qwen3-max-2026-01-23", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-26", - "last_updated": "2025-11-25", + "knowledge": "2025-04", + "release_date": "2026-01-23", + "last_updated": "2026-01-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "siliconflow/deepseek-v3.2": { - "id": "siliconflow/deepseek-v3.2", - "name": "siliconflow/deepseek-v3.2", - "family": "deepseek", + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 163840, + "context": 1000000, "output": 65536 }, "cost": { - "input": 0.27, - "output": 0.42 - } - }, - "siliconflow/deepseek-r1-0528": { - "id": "siliconflow/deepseek-r1-0528", - "name": "siliconflow/deepseek-r1-0528", - "family": "deepseek-thinking", - "attachment": false, + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, + "context": 262144, "output": 32768 }, "cost": { - "input": 0.5, - "output": 2.18 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "siliconflow/deepseek-v3.1-terminus": { - "id": "siliconflow/deepseek-v3.1-terminus", - "name": "siliconflow/deepseek-v3.1-terminus", - "family": "deepseek", + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 196608, + "output": 24576 }, "cost": { - "input": 0.27, - "output": 1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3 Coder Plus", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, + "context": 262144, "output": 65536 }, "cost": { - "input": 1, - "output": 5 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 65536 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } } } }, - "firepass": { - "id": "firepass", - "env": ["FIREPASS_API_KEY"], + "abacus": { + "id": "abacus", + "env": ["ABACUS_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.fireworks.ai/inference/v1/", - "name": "Fireworks (Firepass)", - "doc": "https://docs.fireworks.ai/firepass", + "api": "https://routellm.abacus.ai/v1", + "name": "Abacus", + "doc": "https://abacus.ai/help/api", "models": { - "accounts/fireworks/routers/kimi-k2p6-turbo": { - "id": "accounts/fireworks/routers/kimi-k2p6-turbo", - "name": "Kimi K2.6 Turbo", - "family": "kimi-thinking", - "attachment": false, + "o3": { + "id": "o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 2, + "output": 8, + "cache_read": 0.5 } - } - } - }, - "minimax-cn-coding-plan": { - "id": "minimax-cn-coding-plan", - "env": ["MINIMAX_API_KEY"], - "npm": "@ai-sdk/anthropic", - "api": "https://api.minimaxi.com/anthropic/v1", - "name": "MiniMax Coding Plan (minimaxi.com)", - "doc": "https://platform.minimaxi.com/docs/coding-plan/intro", - "models": { - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax-M2", - "family": "minimax", - "attachment": false, + }, + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, + "route-llm": { + "id": "route-llm", + "name": "RouteLLM", + "description": "RouteLLM routes prompts to an appropriate Abacus-backed text-generation model", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2024-01-01", + "last_updated": "2026-07-10", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 3, + "output": 15 } }, - "MiniMax-M2.7": { - "id": "MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, - "reasoning": true, + "grok-code-fast-1": { + "id": "grok-code-fast-1", + "name": "Grok Code Fast 1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2025-09-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.2, + "output": 1.5 } }, - "MiniMax-M2.7-highspeed": { - "id": "MiniMax-M2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", - "family": "minimax", - "attachment": false, + "gpt-5.3-codex-xhigh": { + "id": "gpt-5.3-codex-xhigh", + "name": "GPT-5.3 Codex XHigh", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1.75, + "output": 14 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", - "name": "MiniMax-M2.1", - "family": "minimax", + "llama-3.3-70b-versatile": { + "id": "llama-3.3-70b-versatile", + "name": "Llama 3.3 70B Versatile", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.59, + "output": 0.79 } }, - "MiniMax-M2.5-highspeed": { - "id": "MiniMax-M2.5-highspeed", - "name": "MiniMax-M2.5-highspeed", - "family": "minimax", - "attachment": false, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } - } - } - }, - "jiekou": { - "id": "jiekou", - "env": ["JIEKOU_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.jiekou.ai/openai", - "name": "Jiekou.AI", - "doc": "https://docs.jiekou.ai/docs/support/quickstart?utm_source=github_models.dev", - "models": { - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "gpt-5.1-codex-max", - "family": "gpt-codex", + }, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32103,76 +73216,86 @@ "open_weights": false, "limit": { "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 1.125, - "output": 9 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "grok-4-1-fast-reasoning", - "family": "grok", + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.18, - "output": 0.45 + "input": 1, + "output": 5 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "claude-opus-4-5-20251101", - "family": "claude-opus", + "grok-4.3": { + "id": "grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 65536 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 4.5, - "output": 22.5 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "gemini-2.5-flash-lite-preview-09-2025", - "family": "gemini-flash-lite", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": false, @@ -32181,221 +73304,269 @@ "output": 65536 }, "cost": { - "input": 0.09, - "output": 0.36 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03 } }, - "gpt-5.2-pro": { - "id": "gpt-5.2-pro", - "name": "gpt-5.2-pro", - "family": "gpt-pro", + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 18.9, - "output": 151.2 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "gemini-3-flash-preview", - "family": "gemini-flash", + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4 + } + }, + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 3 + "input": 1.2, + "output": 6 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "gpt-5-mini", - "family": "gpt-mini", + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.225, - "output": 1.8 + "input": 1.5, + "output": 9, + "cache_read": 0.15 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "gpt-5-nano", - "family": "gpt-nano", + "o3-pro": { + "id": "o3-pro", + "name": "o3-pro", + "description": "High-effort o3 tier for difficult technical reasoning and careful answers", + "family": "o-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-06-10", + "last_updated": "2025-06-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.045, - "output": 0.36 + "input": 20, + "output": 40 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "gemini-3-pro-preview", - "family": "gemini-pro", + "muse-spark-1.1": { + "id": "muse-spark-1.1", + "name": "Muse Spark 1.1", + "description": "Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.", + "family": "muse", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-04-08", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "output": 32000 }, "cost": { - "input": 1.8, - "output": 10.8 + "input": 1.25, + "output": 4.25, + "cache_read": 0.15 } }, - "gemini-2.5-flash-preview-05-20": { - "id": "gemini-2.5-flash-preview-05-20", - "name": "gemini-2.5-flash-preview-05-20", - "family": "gemini-flash", + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 200000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.135, - "output": 3.15 + "input": 15, + "output": 75 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "claude-sonnet-4-5-20250929", - "family": "claude-sonnet", + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2.7, - "output": 13.5 + "input": 5, + "output": 25 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "gemini-2.5-pro", - "family": "gemini-pro", + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 1.125, - "output": 9 + "input": 3, + "output": 15 } }, "grok-4-1-fast-non-reasoning": { "id": "grok-4-1-fast-non-reasoning", - "name": "grok-4-1-fast-non-reasoning", + "name": "Grok 4.1 Fast (Non-Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", "family": "grok", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-11-17", + "last_updated": "2025-11-17", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32403,24 +73574,27 @@ "open_weights": false, "limit": { "context": 2000000, - "output": 2000000 + "output": 16384 }, "cost": { - "input": 0.18, - "output": 0.45 + "input": 0.2, + "output": 0.5 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "gpt-5.2", - "family": "gpt", + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32428,24 +73602,28 @@ "open_weights": false, "limit": { "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 1.575, - "output": 12.6 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o", + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-05", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32453,74 +73631,137 @@ "open_weights": false, "limit": { "context": 200000, - "output": 100000 + "output": 64000 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 5, + "output": 25 } }, - "gemini-2.5-pro-preview-06-05": { - "id": "gemini-2.5-pro-preview-06-05", - "name": "gemini-2.5-pro-preview-06-05", - "family": "gemini-pro", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 200000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1.125, - "output": 9 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "gemini-2.5-flash-lite-preview-06-17": { - "id": "gemini-2.5-flash-lite-preview-06-17", - "name": "gemini-2.5-flash-lite-preview-06-17", - "family": "gemini-flash-lite", + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.25, + "output": 10 + } + }, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25 + } + }, + "claude-opus-4-20250514": { + "id": "claude-opus-4-20250514", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-05-14", + "last_updated": "2025-05-14", "modalities": { - "input": ["text", "video", "image", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.09, - "output": 0.36 + "input": 15, + "output": 75 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "gpt-5.2-codex", - "family": "gpt-codex", + "gpt-5.3-chat-latest": { + "id": "gpt-5.3-chat-latest", + "name": "GPT-5.3 Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32535,42 +73776,159 @@ "output": 14 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "gemini-2.5-flash", + "gemini-3-pro-image-preview": { + "id": "gemini-3-pro-image-preview", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 65536, + "output": 32768 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2 + } + }, + "gemini-2.5-flash-image": { + "id": "gemini-2.5-flash-image", + "name": "Nano Banana", + "description": "Nano Banana image model for fast generation, edits, and character-consistent assets", "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 32768 + }, + "cost": { + "input": 0.3, + "output": 30 + } + }, + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 + } + }, + "kimi-k2-turbo-preview": { + "id": "kimi-k2-turbo-preview", + "name": "Kimi K2 Turbo Preview", + "description": "Fast Kimi model for responsive chat, coding help, and agent loops", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-07-08", + "last_updated": "2025-07-08", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 256000, + "output": 8192 }, "cost": { - "input": 0.27, - "output": 2.25 + "input": 0.15, + "output": 8 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "gpt-5.1-codex-mini", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 + } + }, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32578,24 +73936,27 @@ "open_weights": false, "limit": { "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.225, - "output": 1.8 + "input": 1.75, + "output": 14, + "cache_read": 0.18 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "grok-code-fast-1", + "grok-4-0709": { + "id": "grok-4-0709", + "name": "Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32603,225 +73964,247 @@ "open_weights": false, "limit": { "context": 256000, - "output": 256000 + "output": 16384 }, "cost": { - "input": 0.18, - "output": 1.35 + "input": 3, + "output": 15 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "gpt-5.1", - "family": "gpt", + "grok-4.5": { + "id": "grok-4.5", + "name": "Grok 4.5", + "description": "xAI's latest Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02", - "last_updated": "2026-02", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 500000, + "output": 32768 }, "cost": { - "input": 1.125, - "output": 9 + "input": 2, + "output": 6 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "grok-4-fast-reasoning", - "family": "grok", + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.18, - "output": 0.45 + "input": 1, + "output": 6, + "cache_read": 0.1 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o", + "claude-sonnet-4-20250514": { + "id": "claude-sonnet-4-20250514", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-05-14", + "last_updated": "2025-05-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 3, + "output": 15 } }, - "grok-4-0709": { - "id": "grok-4-0709", - "name": "grok-4-0709", - "family": "grok", - "attachment": true, - "reasoning": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, + "knowledge": "2025-01", "release_date": "2026-01", "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 2.7, - "output": 13.5 + "input": 0.6, + "output": 3 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "gpt-5-codex", - "family": "gpt-codex", + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 1.125, - "output": 9 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "claude-opus-4-1-20250805", - "family": "claude-opus", + "gpt-5.1-chat-latest": { + "id": "gpt-5.1-chat-latest", + "name": "GPT-5.1 Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 13.5, - "output": 67.5 + "input": 1.25, + "output": 10 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "claude-haiku-4-5-20251001", - "family": "claude-haiku", + "gpt-5.2-chat-latest": { + "id": "gpt-5.2-chat-latest", + "name": "GPT-5.2 Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-09-30", + "release_date": "2026-01-01", + "last_updated": "2026-01-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 20000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.9, - "output": 4.5 + "input": 1.75, + "output": 14 } }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "claude-sonnet-4-20250514", - "family": "claude-sonnet", + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 2.7, - "output": 13.5 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "claude-opus-4-6", - "family": "claude-opus", + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02", - "last_updated": "2026-02", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32832,95 +74215,104 @@ "output": 128000 }, "cost": { - "input": 5, - "output": 25 + "input": 10, + "output": 50 } }, - "o3": { - "id": "o3", - "name": "o3", + "gpt-4o-2024-11-20": { + "id": "gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-10", + "release_date": "2024-11-20", + "last_updated": "2024-11-20", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "output": 16384 }, "cost": { - "input": 10, - "output": 40 + "input": 2.5, + "output": 10 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "gpt-5-pro", - "family": "gpt-pro", + "claude-3-7-sonnet-20250219": { + "id": "claude-3-7-sonnet-20250219", + "name": "Claude Sonnet 3.7", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-10-31", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 272000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 13.5, - "output": 108 + "input": 3, + "output": 15 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "gemini-2.5-flash-lite", - "family": "gemini-flash-lite", + "gemini-3.1-flash-image-preview": { + "id": "gemini-3.1-flash-image-preview", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text", "image", "video", "audio"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 65535 + "output": 32768 }, "cost": { - "input": 0.09, - "output": 0.36 + "input": 0.5, + "output": 3 } }, - "gpt-5-chat-latest": { - "id": "gpt-5-chat-latest", - "name": "gpt-5-chat-latest", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -32931,671 +74323,775 @@ "output": 128000 }, "cost": { - "input": 1.125, - "output": 9 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "claude-opus-4-20250514", - "family": "claude-opus", + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 13.5, - "output": 67.5 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "gpt-5.1-codex", - "family": "gpt-codex", + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.125, - "output": 9 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "grok-4-fast-non-reasoning", - "family": "grok", + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.18, - "output": 0.45 + "input": 5, + "output": 25 } }, - "deepseek/deepseek-v3-0324": { - "id": "deepseek/deepseek-v3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", + "qwen-2.5-coder-32b": { + "id": "qwen-2.5-coder-32b", + "name": "Qwen 2.5 Coder 32B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2024-11-11", + "last_updated": "2024-11-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 163840 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.28, - "output": 1.14 + "input": 0.79, + "output": 0.79 } }, - "deepseek/deepseek-v3.1": { - "id": "deepseek/deepseek-v3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": false, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.27, - "output": 1 + "input": 3, + "output": 15 } }, - "deepseek/deepseek-r1-0528": { - "id": "deepseek/deepseek-r1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek-thinking", - "attachment": false, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.7, - "output": 2.5 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "zai-org/glm-4.7": { - "id": "zai-org/glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, - "reasoning": true, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "zai-org/glm-4.5": { - "id": "zai-org/glm-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "zai-org/glm-4.5v": { - "id": "zai-org/glm-4.5v", - "name": "GLM 4.5V", - "family": "glmv", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 3, + "output": 15 } }, - "zai-org/glm-4.7-flash": { - "id": "zai-org/glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm", - "attachment": false, - "reasoning": true, + "grok-4-fast-non-reasoning": { + "id": "grok-4-fast-non-reasoning", + "name": "Grok 4 Fast (Non-Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 2000000, + "output": 16384 }, "cost": { - "input": 0.07, - "output": 0.4 + "input": 0.2, + "output": 0.5 } }, - "minimaxai/minimax-m1-80k": { - "id": "minimaxai/minimax-m1-80k", - "name": "MiniMax M1", - "family": "minimax", - "attachment": false, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 40000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "xiaomimimo/mimo-v2-flash": { - "id": "xiaomimimo/mimo-v2-flash", - "name": "XiaomiMiMo/MiMo-V2-Flash", + "mimo-v2-pro": { + "id": "mimo-v2-pro", + "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 1048576, "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 3, + "cache_read": 0.2 } }, - "baidu/ernie-4.5-vl-424b-a47b": { - "id": "baidu/ernie-4.5-vl-424b-a47b", - "name": "ERNIE 4.5 VL 424B A47B", - "family": "ernie", + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 123000, - "output": 16000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.42, - "output": 1.25 + "input": 0.15, + "output": 0.6 } }, - "baidu/ernie-4.5-300b-a47b-paddle": { - "id": "baidu/ernie-4.5-300b-a47b-paddle", - "name": "ERNIE 4.5 300B A47B", - "family": "ernie", - "attachment": false, - "reasoning": false, + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 123000, - "output": 12000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "Minimax M2.1", - "family": "minimax", + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "qwen/qwen3-235b-a22b-instruct-2507": { - "id": "qwen/qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gemini-3.1-flash-lite-preview": { + "id": "gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.8 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "cache_write": 1 } }, - "qwen/qwen3-32b-fp8": { - "id": "qwen/qwen3-32b-fp8", - "name": "Qwen3 32B", - "family": "qwen", - "attachment": false, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 20000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.45 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "qwen/qwen3-235b-a22b-thinking-2507": { - "id": "qwen/qwen3-235b-a22b-thinking-2507", - "name": "Qwen3 235B A22b Thinking 2507", - "family": "qwen", - "attachment": false, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 3 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "qwen/qwen3-next-80b-a3b-thinking": { - "id": "qwen/qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", - "family": "qwen", - "attachment": false, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 1.5 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "qwen/qwen3-next-80b-a3b-instruct": { - "id": "qwen/qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "family": "qwen", - "attachment": false, + "meta-llama/Meta-Llama-3.3-70B-Instruct": { + "id": "meta-llama/Meta-Llama-3.3-70B-Instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 1.5 + "input": 0.59, + "output": 0.79 } }, - "qwen/qwen3-30b-a3b-fp8": { - "id": "qwen/qwen3-30b-a3b-fp8", - "name": "Qwen3 30B A3B", - "family": "qwen", - "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": false, + "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { + "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", + "name": "Llama 4 Maverick 17B Instruct", + "description": "Open multimodal Llama for strong reasoning with efficient everyday serving", + "family": "llama", + "attachment": true, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 20000 + "context": 1048576, + "output": 8192 }, "cost": { - "input": 0.09, - "output": 0.45 + "input": 0.14, + "output": 0.59 } }, - "qwen/qwen3-coder-next": { - "id": "qwen/qwen3-coder-next", - "name": "qwen/qwen3-coder-next", - "family": "qwen", + "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo": { + "id": "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + "name": "Llama 3.1 405B Instruct Turbo", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02", - "last_updated": "2026-02", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 3.5, + "output": 3.5 } }, - "qwen/qwen3-coder-480b-a35b-instruct": { - "id": "qwen/qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", + "meta-llama/Meta-Llama-3.1-8B-Instruct": { + "id": "meta-llama/Meta-Llama-3.1-8B-Instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.29, - "output": 1.2 + "input": 0.02, + "output": 0.05 } }, - "qwen/qwen3-235b-a22b-fp8": { - "id": "qwen/qwen3-235b-a22b-fp8", - "name": "Qwen3 235B A22B", - "family": "qwen", - "attachment": false, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, - "structured_output": false, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 20000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 131072 }, "cost": { - "input": 0.6, - "output": 3 + "input": 0.14, + "output": 0.4 } }, - "moonshotai/kimi-k2-instruct": { - "id": "moonshotai/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", - "attachment": false, - "reasoning": false, + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 8192 }, "cost": { - "input": 0.57, - "output": 2.3 + "input": 0.32, + "output": 3.2 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "Qwen/Qwen3-Coder-480B-A35B-Instruct": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "name": "Qwen3-Coder 480B-A35B Instruct", + "description": "Open Qwen coding heavyweight for repository reasoning and agentic engineering", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] @@ -33603,119 +75099,103 @@ "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.29, + "output": 1.2 } - } - } - }, - "bailing": { - "id": "bailing", - "env": ["BAILING_API_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.tbox.cn/api/llm/v1/chat/completions", - "name": "Bailing", - "doc": "https://alipaytbox.yuque.com/sxs0ba/ling/intro", - "models": { - "Ring-1T": { - "id": "Ring-1T", - "name": "Ring-1T", - "family": "ring", + }, + "Qwen/QwQ-32B": { + "id": "Qwen/QwQ-32B", + "name": "QwQ 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-10", - "last_updated": "2025-10", + "release_date": "2024-11-28", + "last_updated": "2024-11-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.57, - "output": 2.29 + "input": 0.4, + "output": 0.4 } }, - "Ling-1T": { - "id": "Ling-1T", - "name": "Ling-1T", - "family": "ling", + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-10", - "last_updated": "2025-10", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 262144, + "output": 8192 }, "cost": { - "input": 0.57, - "output": 2.29 + "input": 0.13, + "output": 0.6 } - } - } - }, - "iflowcn": { - "id": "iflowcn", - "env": ["IFLOW_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://apis.iflow.cn/v1", - "name": "iFlow", - "doc": "https://platform.iflow.cn/en/docs", - "models": { - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3-Coder-Plus", + }, + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.09, + "output": 0.29 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3-32B", + "Qwen/Qwen2.5-72B-Instruct": { + "id": "Qwen/Qwen2.5-72B-Instruct", + "name": "Qwen 2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { "input": ["text"], "output": ["text"] @@ -33723,24 +75203,26 @@ "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.11, + "output": 0.38 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] @@ -33748,174 +75230,188 @@ "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.08, + "output": 0.44 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3-Max", - "family": "qwen", + "zai-org/GLM-4.6": { + "id": "zai-org/GLM-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.2 } }, - "qwen3-235b-a22b-instruct": { - "id": "qwen3-235b-a22b-instruct", - "name": "Qwen3-235B-A22B-Instruct", - "family": "qwen", + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 3.2 } }, - "qwen3-235b-a22b-thinking-2507": { - "id": "qwen3-235b-a22b-thinking-2507", - "name": "Qwen3-235B-A22B-Thinking", - "family": "qwen", + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.2 } }, - "kimi-k2-0905": { - "id": "kimi-k2-0905", - "name": "Kimi-K2-0905", - "family": "kimi", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "GLM-4.6", + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2025-11-13", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "qwen3-vl-plus": { - "id": "qwen3-vl-plus", - "name": "Qwen3-VL-Plus", - "family": "qwen", - "attachment": true, - "reasoning": false, + "zai-org/GLM-4.5": { + "id": "zai-org/GLM-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 131072, + "output": 96000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.2 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek-V3.2-Exp", - "family": "deepseek", + "deepseek-ai/DeepSeek-R1": { + "id": "deepseek-ai/DeepSeek-R1", + "name": "DeepSeek R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] @@ -33923,24 +75419,25 @@ "open_weights": true, "limit": { "context": 128000, - "output": 64000 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 7 } }, - "qwen3-235b": { - "id": "qwen3-235b", - "name": "Qwen3-235B-A22B", - "family": "qwen", + "deepseek-ai/DeepSeek-V3.1-Terminus": { + "id": "deepseek-ai/DeepSeek-V3.1-Terminus", + "name": "DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { "input": ["text"], "output": ["text"] @@ -33948,74 +75445,83 @@ "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.27, + "output": 1 } }, - "kimi-k2": { - "id": "kimi-k2", - "name": "Kimi-K2", - "family": "kimi", + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.14, + "output": 0.28, + "cache_read": 0.03 } }, - "qwen3-max-preview": { - "id": "qwen3-max-preview", - "name": "Qwen3-Max-Preview", - "family": "qwen", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 1.74, + "output": 3.48, + "cache_read": 0.15 } }, - "deepseek-v3": { - "id": "deepseek-v3", - "name": "DeepSeek-V3", + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "family": "deepseek", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-26", - "last_updated": "2024-12-26", + "release_date": "2025-06-15", + "last_updated": "2025-06-15", "modalities": { "input": ["text"], "output": ["text"] @@ -34023,193 +75529,197 @@ "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.27, + "output": 0.4 } - } - } - }, - "v0": { - "id": "v0", - "env": ["V0_API_KEY"], - "npm": "@ai-sdk/vercel", - "name": "v0", - "doc": "https://sdk.vercel.ai/providers/ai-sdk-providers/vercel", - "models": { - "v0-1.5-lg": { - "id": "v0-1.5-lg", - "name": "v0-1.5-lg", - "family": "v0", + }, + "MiniMaxAI/MiniMax-M3": { + "id": "MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-06-09", - "last_updated": "2025-06-09", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 512000, - "output": 32000 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 15, - "output": 75 + "input": 0.3, + "output": 1.2 } }, - "v0-1.0-md": { - "id": "v0-1.0-md", - "name": "v0-1.0-md", - "family": "v0", - "attachment": true, + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 3, - "output": 15 + "input": 0.3, + "output": 1.2 } }, - "v0-1.5-md": { - "id": "v0-1.5-md", - "name": "v0-1.5-md", - "family": "v0", - "attachment": true, + "deepseek/deepseek-v3.1": { + "id": "deepseek/deepseek-v3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-06-09", - "last_updated": "2025-06-09", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 8192 }, "cost": { - "input": 3, - "output": 15 + "input": 0.55, + "output": 1.66 } } } }, - "huggingface": { - "id": "huggingface", - "env": ["HF_TOKEN"], + "cloudferro-sherlock": { + "id": "cloudferro-sherlock", + "env": ["CLOUDFERRO_SHERLOCK_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://router.huggingface.co/v1", - "name": "Hugging Face", - "doc": "https://huggingface.co/docs/inference-providers", + "api": "https://api-sherlock.cloudferro.com/openai/v1/", + "name": "CloudFerro Sherlock", + "doc": "https://docs.sherlock.cloudferro.com/", "models": { - "Qwen/Qwen3.5-397B-A17B": { - "id": "Qwen/Qwen3.5-397B-A17B", - "name": "Qwen3.5-397B-A17B", - "family": "qwen", - "attachment": true, - "reasoning": true, + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "knowledge": "2024-10-09", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 70000, + "output": 70000 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 2.92, + "output": 2.92 } }, - "Qwen/Qwen3-Coder-Next": { - "id": "Qwen/Qwen3-Coder-Next", - "name": "Qwen3-Coder-Next", - "family": "qwen", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "OpenAI GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-03", - "last_updated": "2026-02-03", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131000, + "output": 131000 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 2.92, + "output": 2.92 } }, - "Qwen/Qwen3-Next-80B-A3B-Instruct": { - "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "name": "Qwen3-Next-80B-A3B-Instruct", - "family": "qwen", + "speakleash/Bielik-11B-v3.0-Instruct": { + "id": "speakleash/Bielik-11B-v3.0-Instruct", + "name": "Bielik 11B v3.0 Instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-11", - "last_updated": "2025-09-11", + "knowledge": "2025-03", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 66536 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.67, + "output": 0.67 } }, - "Qwen/Qwen3-Embedding-8B": { - "id": "Qwen/Qwen3-Embedding-8B", - "name": "Qwen 3 Embedding 8B", - "family": "qwen", + "speakleash/Bielik-11B-v2.6-Instruct": { + "id": "speakleash/Bielik-11B-v2.6-Instruct", + "name": "Bielik 11B v2.6 Instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-03", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] @@ -34217,395 +75727,410 @@ "open_weights": true, "limit": { "context": 32000, - "output": 4096 + "output": 32000 }, "cost": { - "input": 0.01, - "output": 0 + "input": 0.67, + "output": 0.67 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3-235B-A22B-Thinking-2507", - "family": "qwen", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "knowledge": "2026-01", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 131072 + "context": 196000, + "input": 180000, + "output": 16000 }, "cost": { "input": 0.3, - "output": 3 + "output": 1.2 } - }, - "Qwen/Qwen3-Next-80B-A3B-Thinking": { - "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "name": "Qwen3-Next-80B-A3B-Thinking", - "family": "qwen", + } + } + }, + "ollama-cloud": { + "id": "ollama-cloud", + "env": ["OLLAMA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://ollama.com/v1", + "name": "Ollama Cloud", + "doc": "https://docs.ollama.com/cloud", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "deepseek-v4-flash", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-11", - "last_updated": "2025-09-11", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 131072 - }, - "cost": { - "input": 0.3, - "output": 2 + "context": 1048576, + "output": 1048576 } }, - "Qwen/Qwen3-Embedding-4B": { - "id": "Qwen/Qwen3-Embedding-4B", - "name": "Qwen 3 Embedding 4B", - "family": "qwen", + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "minimax-m2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "knowledge": "2025-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 2048 - }, - "cost": { - "input": 0.01, - "output": 0 + "context": 204800, + "output": 131072 } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen3-Coder-480B-A35B-Instruct", - "family": "qwen", - "attachment": false, + "devstral-small-2:24b": { + "id": "devstral-small-2:24b", + "name": "devstral-small-2:24b", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2025-12-09", + "last_updated": "2026-01-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 66536 - }, - "cost": { - "input": 2, - "output": 2 + "output": 262144 } }, - "zai-org/GLM-4.7-Flash": { - "id": "zai-org/GLM-4.7-Flash", - "name": "GLM-4.7-Flash", + "glm-4.7": { + "id": "glm-4.7", + "name": "glm-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "release_date": "2025-12-22", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 - }, - "cost": { - "input": 0, - "output": 0 + "context": 202752, + "output": 131072 } }, - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", - "name": "GLM-4.7", - "family": "glm", + "cogito-2.1:671b": { + "id": "cogito-2.1:671b", + "name": "cogito-2.1:671b", + "description": "Legacy model retained for compatibility with older integrations", + "family": "cogito", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2025-11-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 163840, + "output": 32000 }, - "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11 - } + "status": "deprecated" }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", - "name": "GLM-5.1", - "family": "glm", + "minimax-m2.1": { + "id": "minimax-m2.1", + "name": "minimax-m2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-03", - "last_updated": "2026-04-03", + "release_date": "2025-12-23", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, + "context": 204800, "output": 131072 - }, - "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", - "family": "glm", + "gpt-oss:120b": { + "id": "gpt-oss:120b", + "name": "gpt-oss:120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2025-08-05", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 - }, - "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "context": 131072, + "output": 32768 } }, - "XiaomiMiMo/MiMo-V2-Flash": { - "id": "XiaomiMiMo/MiMo-V2-Flash", - "name": "MiMo-V2-Flash", - "family": "mimo", + "nemotron-3-nano:30b": { + "id": "nemotron-3-nano:30b", + "name": "nemotron-3-nano:30b", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "release_date": "2025-12-15", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 4096 + "context": 1048576, + "output": 131072 + } + }, + "ministral-3:8b": { + "id": "ministral-3:8b", + "name": "ministral-3:8b", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, + "reasoning": false, + "tool_call": true, + "release_date": "2024-12-01", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "cost": { - "input": 0.1, - "output": 0.3 + "open_weights": true, + "limit": { + "context": 262144, + "output": 128000 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek-R1-0528", - "family": "deepseek-thinking", + "rnj-1:8b": { + "id": "rnj-1:8b", + "name": "rnj-1:8b", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "rnj", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "release_date": "2025-12-06", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 163840 - }, - "cost": { - "input": 3, - "output": 5 + "context": 32768, + "output": 4096 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek-V3.2", - "family": "deepseek", - "attachment": false, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "kimi-k2.7-code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 65536 - }, - "cost": { - "input": 0.28, - "output": 0.4 + "context": 262144, + "output": 262144 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "Kimi-K2-Thinking", - "family": "kimi-thinking", + "glm-5.1": { + "id": "glm-5.1", + "name": "glm-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "release_date": "2026-03-27", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 - }, - "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "context": 202752, + "output": 131072 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "Kimi-K2.6", - "family": "kimi", - "attachment": true, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "deepseek-v4-pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 - }, - "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "context": 1048576, + "output": 1048576 } }, - "moonshotai/Kimi-K2-Instruct": { - "id": "moonshotai/Kimi-K2-Instruct", - "name": "Kimi-K2-Instruct", - "family": "kimi", + "glm-4.6": { + "id": "glm-4.6", + "name": "glm-4.6", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-14", - "last_updated": "2025-07-14", + "release_date": "2025-09-29", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 202752, + "output": 131072 }, - "cost": { - "input": 1, - "output": 3 - } + "status": "deprecated" }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi-K2-Instruct-0905", - "family": "kimi", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "kimi-k2-thinking", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-04", - "last_updated": "2025-09-04", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] @@ -34613,113 +76138,129 @@ "open_weights": true, "limit": { "context": 262144, - "output": 16384 + "output": 262144 }, - "cost": { - "input": 1, - "output": 3 - } + "status": "deprecated" }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi-K2.5", - "family": "kimi", - "attachment": true, + "nemotron-3-super": { + "id": "nemotron-3-super", + "name": "nemotron-3-super", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-01", - "last_updated": "2026-01-01", + "release_date": "2026-03-11", + "last_updated": "2026-03-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 65536 + } + }, + "ministral-3:14b": { + "id": "ministral-3:14b", + "name": "ministral-3:14b", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, + "reasoning": false, + "tool_call": true, + "release_date": "2024-12-01", + "last_updated": "2026-01-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "open_weights": true, + "limit": { + "context": 262144, + "output": 128000 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 976000, "output": 131072 - }, - "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 } - }, - "MiniMaxAI/MiniMax-M2.7": { - "id": "MiniMaxAI/MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, + }, + "minimax-m3": { + "id": "minimax-m3", + "name": "minimax-m3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax-m3", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-01", + "release_date": "2026-05-31", + "last_updated": "2026-05-31", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 512000, "output": 131072 - }, - "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 } }, - "MiniMaxAI/MiniMax-M2.1": { - "id": "MiniMaxAI/MiniMax-M2.1", - "name": "MiniMax-M2.1", + "minimax-m2": { + "id": "minimax-m2", + "name": "minimax-m2", + "description": "Legacy model retained for compatibility with older integrations", "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-10", - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2025-10-23", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] @@ -34727,1098 +76268,1052 @@ "open_weights": true, "limit": { "context": 204800, - "output": 131072 + "output": 128000 }, - "cost": { - "input": 0.3, - "output": 1.2 - } + "status": "deprecated" }, - "deepseek-ai/DeepSeek-V4-Pro": { - "id": "deepseek-ai/DeepSeek-V4-Pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "qwen3-next:80b": { + "id": "qwen3-next:80b", + "name": "qwen3-next:80b", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-09-15", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 262144, + "output": 32768 }, - "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 - } - } - } - }, - "zenmux": { - "id": "zenmux", - "env": ["ZENMUX_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://zenmux.ai/api/v1", - "name": "ZenMux", - "doc": "https://docs.zenmux.ai", - "models": { - "deepseek/deepseek-chat": { - "id": "deepseek/deepseek-chat", - "name": "DeepSeek-V3.2 (Non-thinking Mode)", + "status": "deprecated" + }, + "qwen3-coder:480b": { + "id": "qwen3-coder:480b", + "name": "qwen3-coder:480b", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2025-07-22", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 - }, - "cost": { - "input": 0.28, - "output": 0.42, - "cache_read": 0.03 + "context": 262144, + "output": 65536 } }, - "deepseek/deepseek-v3.2-exp": { - "id": "deepseek/deepseek-v3.2-exp", - "name": "DeepSeek-V3.2-Exp", + "kimi-k2:1t": { + "id": "kimi-k2:1t", + "name": "kimi-k2:1t", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2024-10", + "release_date": "2025-07-11", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163000, - "output": 64000 + "context": 262144, + "output": 262144 }, - "cost": { - "input": 0.22, - "output": 0.33 - } + "status": "deprecated" }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "DeepSeek V3.2", + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "minimax-m2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-05", - "last_updated": "2025-12-05", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 - }, - "cost": { - "input": 0.28, - "output": 0.43 + "context": 196608, + "output": 196608 } }, - "inclusionai/ring-1t": { - "id": "inclusionai/ring-1t", - "name": "Ring-1T", + "deepseek-v3.1:671b": { + "id": "deepseek-v3.1:671b", + "name": "deepseek-v3.1:671b", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-10-12", - "last_updated": "2025-10-12", + "release_date": "2025-08-21", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 - }, - "cost": { - "input": 0.56, - "output": 2.24, - "cache_read": 0.11 + "context": 163840, + "output": 163840 } }, - "inclusionai/ling-1t": { - "id": "inclusionai/ling-1t", - "name": "Ling-1T", - "attachment": false, - "reasoning": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "kimi-k2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-10-09", - "last_updated": "2025-10-09", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 - }, - "cost": { - "input": 0.56, - "output": 2.24, - "cache_read": 0.11 + "context": 262144, + "output": 262144 } }, - "stepfun/step-3.5-flash-free": { - "id": "stepfun/step-3.5-flash-free", - "name": "Step 3.5 Flash (Free)", - "attachment": false, + "gemma4:31b": { + "id": "gemma4:31b", + "name": "gemma4:31b", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 - }, - "cost": { - "input": 0, - "output": 0 + "context": 262144, + "output": 262144 } }, - "stepfun/step-3.5-flash": { - "id": "stepfun/step-3.5-flash", - "name": "Step 3.5 Flash", - "attachment": false, + "ministral-3:3b": { + "id": "ministral-3:3b", + "name": "ministral-3:3b", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "release_date": "2024-10-22", + "last_updated": "2026-01-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 - }, - "cost": { - "input": 0.1, - "output": 0.3 + "context": 262144, + "output": 128000 } }, - "stepfun/step-3": { - "id": "stepfun/step-3", - "name": "Step-3", + "gemma3:12b": { + "id": "gemma3:12b", + "name": "gemma3:12b", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "reasoning": false, + "tool_call": false, + "release_date": "2024-12-01", + "last_updated": "2026-01-19", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "output": 64000 - }, - "cost": { - "input": 0.21, - "output": 0.57 + "context": 131072, + "output": 131072 } }, - "kuaishou/kat-coder-pro-v2": { - "id": "kuaishou/kat-coder-pro-v2", - "name": "KAT-Coder-Pro-V2", - "attachment": false, + "gemma3:4b": { + "id": "gemma3:4b", + "name": "gemma3:4b", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "tool_call": false, + "release_date": "2024-12-01", + "last_updated": "2026-01-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 80000 - }, - "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "context": 131072, + "output": 131072 } }, - "x-ai/grok-4-fast": { - "id": "x-ai/grok-4-fast", - "name": "Grok 4 Fast", + "qwen3.5:397b": { + "id": "qwen3.5:397b", + "name": "qwen3.5:397b", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "interleaved": { + "field": "reasoning_details" + }, + "release_date": "2026-02-15", + "last_updated": "2026-02-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 64000 + "context": 262144, + "output": 65536 + } + }, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "kimi-k2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 } }, - "x-ai/grok-code-fast-1": { - "id": "x-ai/grok-code-fast-1", - "name": "Grok Code Fast 1", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "qwen3-coder-next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "release_date": "2026-02-02", + "last_updated": "2026-02-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 - }, - "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "context": 262144, + "output": 65536 } }, - "x-ai/grok-4.1-fast-non-reasoning": { - "id": "x-ai/grok-4.1-fast-non-reasoning", - "name": "Grok 4.1 Fast Non Reasoning", + "qwen3-vl:235b-instruct": { + "id": "qwen3-vl:235b-instruct", + "name": "qwen3-vl:235b-instruct", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "release_date": "2025-09-22", + "last_updated": "2026-01-19", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 64000 + "context": 262144, + "output": 131072 }, - "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 - } + "status": "deprecated" }, - "x-ai/grok-4": { - "id": "x-ai/grok-4", - "name": "Grok 4", + "mistral-large-3:675b": { + "id": "mistral-large-3:675b", + "name": "mistral-large-3:675b", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "release_date": "2025-12-02", + "last_updated": "2026-01-19", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "context": 262144, + "output": 262144 } }, - "x-ai/grok-4.1-fast": { - "id": "x-ai/grok-4.1-fast", - "name": "Grok 4.1 Fast", + "gemma3:27b": { + "id": "gemma3:27b", + "name": "gemma3:27b", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "reasoning": false, + "tool_call": false, + "release_date": "2025-07-27", + "last_updated": "2026-01-19", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 64000 - }, - "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "context": 131072, + "output": 131072 } }, - "x-ai/grok-4.2-fast": { - "id": "x-ai/grok-4.2-fast", - "name": "Grok 4.2 Fast", + "qwen3-vl:235b": { + "id": "qwen3-vl:235b", + "name": "qwen3-vl:235b", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-09-22", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "output": 32768 }, - "cost": { - "input": 3, - "output": 9 - } + "status": "deprecated" }, - "x-ai/grok-4.2-fast-non-reasoning": { - "id": "x-ai/grok-4.2-fast-non-reasoning", - "name": "Grok 4.2 Fast Non Reasoning", - "attachment": true, - "reasoning": false, + "nemotron-3-ultra": { + "id": "nemotron-3-ultra", + "name": "nemotron-3-ultra", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 - }, - "cost": { - "input": 3, - "output": 9 + "context": 262144, + "output": 128000 } }, - "openai/gpt-5.3-chat": { - "id": "openai/gpt-5.3-chat", - "name": "GPT-5.3 Chat", + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "gemini-3-flash-preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2026-04-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16380 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" - }, - "cost": { - "input": 1.75, - "output": 14 + "context": 1048576, + "output": 65536 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT-5.2-Pro", - "attachment": true, + "gpt-oss:20b": { + "id": "gpt-oss:20b", + "name": "gpt-oss:20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "release_date": "2025-08-05", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" - }, - "cost": { - "input": 21, - "output": 168 + "context": 131072, + "output": 32768 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "attachment": true, + "glm-5": { + "id": "glm-5", + "name": "glm-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "interleaved": { + "field": "reasoning_content" + }, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" - }, - "cost": { - "input": 1.75, - "output": 14 + "context": 202752, + "output": 131072 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "attachment": true, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "deepseek-v3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-01-01", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "release_date": "2025-06-15", + "last_updated": "2026-01-19", "modalities": { - "input": ["image", "text", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" - }, - "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.17 + "context": 163840, + "output": 65536 } }, - "openai/gpt-5.4-mini": { - "id": "openai/gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "attachment": true, + "devstral-2:123b": { + "id": "devstral-2:123b", + "name": "devstral-2:123b", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-12-09", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" - }, - "cost": { - "input": 0.75, - "output": 4.5 + "context": 262144, + "output": 262144 } - }, - "openai/gpt-5.1-chat": { - "id": "openai/gpt-5.1-chat", - "name": "GPT-5.1 Chat", - "attachment": true, + } + } + }, + "cloudflare-ai-gateway": { + "id": "cloudflare-ai-gateway", + "env": ["CLOUDFLARE_API_TOKEN", "CLOUDFLARE_ACCOUNT_ID", "CLOUDFLARE_GATEWAY_ID"], + "npm": "ai-gateway-provider", + "name": "Cloudflare AI Gateway", + "doc": "https://developers.cloudflare.com/ai-gateway/", + "models": { + "workers-ai/@cf/baai/bge-m3": { + "id": "workers-ai/@cf/baai/bge-m3", + "name": "BGE M3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "bge", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["pdf", "image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12 + "input": 0.012, + "output": 0 } }, - "openai/gpt-5.4-nano": { - "id": "openai/gpt-5.4-nano", - "name": "GPT-5.4 Nano", + "workers-ai/@cf/baai/bge-small-en-v1.5": { + "id": "workers-ai/@cf/baai/bge-small-en-v1.5", + "name": "BGE Small EN v1.5", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "bge", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 1.25 + "input": 0.02, + "output": 0 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2025-01-01", - "release_date": "2026-01-15", - "last_updated": "2026-01-15", + "workers-ai/@cf/baai/bge-reranker-base": { + "id": "workers-ai/@cf/baai/bge-reranker-base", + "name": "BGE Reranker Base", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "bge", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-04-09", + "last_updated": "2025-04-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.17 + "input": 0.0031, + "output": 0 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT-5.1-Codex-Mini", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/baai/bge-base-en-v1.5": { + "id": "workers-ai/@cf/baai/bge-base-en-v1.5", + "name": "BGE Base EN v1.5", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "bge", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.03 + "input": 0.067, + "output": 0 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/baai/bge-large-en-v1.5": { + "id": "workers-ai/@cf/baai/bge-large-en-v1.5", + "name": "BGE Large EN v1.5", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "bge", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["image", "text", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12 + "input": 0.2, + "output": 0 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/ai4bharat/indictrans2-en-indic-1B": { + "id": "workers-ai/@cf/ai4bharat/indictrans2-en-indic-1B", + "name": "IndicTrans2 EN-Indic 1B", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "indictrans", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 45, - "output": 225 + "input": 0.34, + "output": 0.34 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5 Codex", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/ibm-granite/granite-4.0-h-micro": { + "id": "workers-ai/@cf/ibm-granite/granite-4.0-h-micro", + "name": "IBM Granite 4.0 H Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "granite", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12 + "input": 0.017, + "output": 0.11 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/huggingface/distilbert-sst-2-int8": { + "id": "workers-ai/@cf/huggingface/distilbert-sst-2-int8", + "name": "DistilBERT SST-2 INT8", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "distilbert", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 3.75, - "output": 18.75 + "input": 0.026, + "output": 0 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", + "workers-ai/@cf/moonshotai/kimi-k2.5": { + "id": "workers-ai/@cf/moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 256000, + "output": 256000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1-Codex", + "workers-ai/@cf/moonshotai/kimi-k2.6": { + "id": "workers-ai/@cf/moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/openai", - "api": "https://zenmux.ai/api/v1" + "context": 256000, + "output": 256000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "z-ai/glm-4.7-flash-free": { - "id": "z-ai/glm-4.7-flash-free", - "name": "GLM 4.7 Flash (Free)", + "workers-ai/@cf/mistral/mistral-7b-instruct-v0.1": { + "id": "workers-ai/@cf/mistral/mistral-7b-instruct-v0.1", + "name": "Mistral 7B Instruct v0.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.11, + "output": 0.19 } }, - "z-ai/glm-5v-turbo": { - "id": "z-ai/glm-5v-turbo", - "name": "GLM 5V Turbo", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "workers-ai/@cf/google/gemma-3-12b-it": { + "id": "workers-ai/@cf/google/gemma-3-12b-it", + "name": "Gemma 3 12B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { - "input": ["text", "image", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.726, - "output": 3.1946, - "cache_read": 0.1743 + "input": 0.35, + "output": 0.56 } }, - "z-ai/glm-4.7": { - "id": "z-ai/glm-4.7", - "name": "GLM 4.7", + "workers-ai/@cf/myshell-ai/melotts": { + "id": "workers-ai/@cf/myshell-ai/melotts", + "name": "MyShell MeloTTS", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "melotts", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.28, - "output": 1.14, - "cache_read": 0.06 + "input": 0, + "output": 0 } }, - "z-ai/glm-5": { - "id": "z-ai/glm-5", - "name": "GLM 5", + "workers-ai/@cf/openai/gpt-oss-120b": { + "id": "workers-ai/@cf/openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.58, - "output": 2.6, - "cache_read": 0.14 + "input": 0.35, + "output": 0.75 } }, - "z-ai/glm-4.7-flashx": { - "id": "z-ai/glm-4.7-flashx", - "name": "GLM 4.7 FlashX", + "workers-ai/@cf/openai/gpt-oss-20b": { + "id": "workers-ai/@cf/openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.07, - "output": 0.42, - "cache_read": 0.01 + "input": 0.2, + "output": 0.3 } }, - "z-ai/glm-4.6v-flash-free": { - "id": "z-ai/glm-4.6v-flash-free", - "name": "GLM 4.6V Flash (Free)", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct": { + "id": "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct", + "name": "Mistral Small 3.1 24B Instruct", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.35, + "output": 0.56 } }, - "z-ai/glm-5.1": { - "id": "z-ai/glm-5.1", - "name": "GLM-5.1", + "workers-ai/@cf/nvidia/nemotron-3-120b-a12b": { + "id": "workers-ai/@cf/nvidia/nemotron-3-120b-a12b", + "name": "Nemotron 3 Super 120B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-03", - "last_updated": "2026-04-03", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.8781, - "output": 3.5126, - "cache_read": 0.1903 + "input": 0.5, + "output": 1.5 } }, - "z-ai/glm-4.6v-flash": { - "id": "z-ai/glm-4.6v-flash", - "name": "GLM 4.6V FlashX", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/pfnet/plamo-embedding-1b": { + "id": "workers-ai/@cf/pfnet/plamo-embedding-1b", + "name": "PLaMo Embedding 1B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "plamo", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.02, - "output": 0.21, - "cache_read": 0.0043 + "input": 0.019, + "output": 0 } }, - "z-ai/glm-4.5": { - "id": "z-ai/glm-4.5", - "name": "GLM 4.5", + "workers-ai/@cf/deepgram/aura-2-es": { + "id": "workers-ai/@cf/deepgram/aura-2-es", + "name": "Deepgram Aura 2 (ES)", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "aura", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] @@ -35826,24 +77321,24 @@ "open_weights": false, "limit": { "context": 128000, - "output": 64000 + "output": 16384 }, "cost": { - "input": 0.35, - "output": 1.54, - "cache_read": 0.07 + "input": 0, + "output": 0 } }, - "z-ai/glm-4.5-air": { - "id": "z-ai/glm-4.5-air", - "name": "GLM 4.5 Air", + "workers-ai/@cf/deepgram/aura-2-en": { + "id": "workers-ai/@cf/deepgram/aura-2-en", + "name": "Deepgram Aura 2 (EN)", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "aura", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] @@ -35851,1164 +77346,1246 @@ "open_weights": false, "limit": { "context": 128000, - "output": 64000 + "output": 16384 }, "cost": { - "input": 0.11, - "output": 0.56, - "cache_read": 0.02 + "input": 0, + "output": 0 } }, - "z-ai/glm-5-turbo": { - "id": "z-ai/glm-5-turbo", - "name": "GLM 5 Turbo", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/deepgram/nova-3": { + "id": "workers-ai/@cf/deepgram/nova-3", + "name": "Deepgram Nova 3", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "nova", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.88, - "output": 3.48 + "input": 0, + "output": 0 } }, - "z-ai/glm-4.6": { - "id": "z-ai/glm-4.6", - "name": "GLM 4.6", + "workers-ai/@cf/zai-org/glm-4.7-flash": { + "id": "workers-ai/@cf/zai-org/glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.35, - "output": 1.54, - "cache_read": 0.07 + "input": 0.06, + "output": 0.4 } }, - "z-ai/glm-4.6v": { - "id": "z-ai/glm-4.6v", - "name": "GLM 4.6V", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/deepseek-ai/deepseek-r1-distill-qwen-32b": { + "id": "workers-ai/@cf/deepseek-ai/deepseek-r1-distill-qwen-32b", + "name": "DeepSeek R1 Distill Qwen 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.42, - "cache_read": 0.03 + "input": 0.5, + "output": 4.88 } }, - "volcengine/doubao-seed-2.0-code": { - "id": "volcengine/doubao-seed-2.0-code", - "name": "Doubao Seed 2.0 Code", - "attachment": true, + "workers-ai/@cf/qwen/qwen3-embedding-0.6b": { + "id": "workers-ai/@cf/qwen/qwen3-embedding-0.6b", + "name": "Qwen3 Embedding 0.6B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 32000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.9, - "output": 4.48 + "input": 0.012, + "output": 0 } }, - "volcengine/doubao-seed-code": { - "id": "volcengine/doubao-seed-code", - "name": "Doubao-Seed-Code", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8": { + "id": "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8", + "name": "Qwen3 30B A3B FP8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-11", - "last_updated": "2025-11-11", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.17, - "output": 1.12, - "cache_read": 0.03 + "input": 0.051, + "output": 0.34 } }, - "volcengine/doubao-seed-2.0-mini": { - "id": "volcengine/doubao-seed-2.0-mini", - "name": "Doubao-Seed-2.0-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/qwen/qwen2.5-coder-32b-instruct": { + "id": "workers-ai/@cf/qwen/qwen2.5-coder-32b-instruct", + "name": "Qwen 2.5 Coder 32B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2026-02-14", - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.03, - "output": 0.28, - "cache_read": 0.01, - "cache_write": 0.0024 + "input": 0.66, + "output": 1 } }, - "volcengine/doubao-seed-2.0-lite": { - "id": "volcengine/doubao-seed-2.0-lite", - "name": "Doubao-Seed-2.0-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/qwen/qwq-32b": { + "id": "workers-ai/@cf/qwen/qwq-32b", + "name": "QwQ 32B", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2026-02-14", - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.09, - "output": 0.51, - "cache_read": 0.02, - "cache_write": 0.0024 + "input": 0.66, + "output": 1 } }, - "volcengine/doubao-seed-1.8": { - "id": "volcengine/doubao-seed-1.8", - "name": "Doubao-Seed-1.8", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/pipecat-ai/smart-turn-v2": { + "id": "workers-ai/@cf/pipecat-ai/smart-turn-v2", + "name": "Pipecat Smart Turn v2", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "smart-turn", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.11, - "output": 0.28, - "cache_read": 0.02, - "cache_write": 0.0024 + "input": 0, + "output": 0 } }, - "volcengine/doubao-seed-2.0-pro": { - "id": "volcengine/doubao-seed-2.0-pro", - "name": "Doubao-Seed-2.0-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/meta/llama-3.1-8b-instruct": { + "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2026-02-14", - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.45, - "output": 2.24, - "cache_read": 0.09, - "cache_write": 0.0024 + "input": 0.28, + "output": 0.8299999999999998 } }, - "baidu/ernie-5.0-thinking-preview": { - "id": "baidu/ernie-5.0-thinking-preview", - "name": "ERNIE 5.0", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/meta/m2m100-1.2b": { + "id": "workers-ai/@cf/meta/m2m100-1.2b", + "name": "M2M100 1.2B", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "m2m", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 64000 + "output": 16384 }, "cost": { - "input": 0.84, - "output": 3.37 + "input": 0.34, + "output": 0.34 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "MiniMax M2.7", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/meta/llama-3.2-1b-instruct": { + "id": "workers-ai/@cf/meta/llama-3.2-1b-instruct", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131070 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3055, - "output": 1.2219 + "input": 0.027, + "output": 0.2 } }, - "minimax/minimax-m2.7-highspeed": { - "id": "minimax/minimax-m2.7-highspeed", - "name": "MiniMax M2.7 highspeed", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/meta/llama-3.2-11b-vision-instruct": { + "id": "workers-ai/@cf/meta/llama-3.2-11b-vision-instruct", + "name": "Llama 3.2 11B Vision Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131070 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.611, - "output": 2.4439 + "input": 0.049, + "output": 0.68 } }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "MiniMax M2", + "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct": { + "id": "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.38 + "input": 0.27, + "output": 0.85 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "MiniMax M2.1", + "workers-ai/@cf/meta/llama-guard-3-8b": { + "id": "workers-ai/@cf/meta/llama-guard-3-8b", + "name": "Llama Guard 3 8B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.38 + "input": 0.48, + "output": 0.03 } }, - "minimax/minimax-m2.5-lightning": { - "id": "minimax/minimax-m2.5-lightning", - "name": "MiniMax M2.5 highspeed", + "workers-ai/@cf/meta/llama-3-8b-instruct-awq": { + "id": "workers-ai/@cf/meta/llama-3-8b-instruct-awq", + "name": "Llama 3 8B Instruct AWQ", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 4.8, - "cache_read": 0.06, - "cache_write": 0.75 + "input": 0.12, + "output": 0.27 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", + "workers-ai/@cf/meta/llama-3.1-8b-instruct-awq": { + "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct-awq", + "name": "Llama 3.1 8B Instruct AWQ", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 0.12, + "output": 0.27 } }, - "qwen/qwen3-coder-plus": { - "id": "qwen/qwen3-coder-plus", - "name": "Qwen3-Coder-Plus", + "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast": { + "id": "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast", + "name": "Llama 3.3 70B Instruct FP8 Fast", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.29, + "output": 2.25 } }, - "qwen/qwen3.5-flash": { - "id": "qwen/qwen3.5-flash", - "name": "Qwen3.5 Flash", - "attachment": true, + "workers-ai/@cf/meta/llama-3-8b-instruct": { + "id": "workers-ai/@cf/meta/llama-3-8b-instruct", + "name": "Llama 3 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1020000, - "output": 1020000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.28, + "output": 0.83 } }, - "qwen/qwen3.6-plus": { - "id": "qwen/qwen3.6-plus", - "name": "Qwen3.6-Plus", + "workers-ai/@cf/meta/llama-3.1-8b-instruct-fp8": { + "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct-fp8", + "name": "Llama 3.1 8B Instruct FP8", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5 - } + "input": 0.15, + "output": 0.29 } }, - "qwen/qwen3-max": { - "id": "qwen/qwen3-max", - "name": "Qwen3-Max-Thinking", + "workers-ai/@cf/meta/llama-2-7b-chat-fp16": { + "id": "workers-ai/@cf/meta/llama-2-7b-chat-fp16", + "name": "Llama 2 7B Chat FP16", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-01-23", - "last_updated": "2026-01-23", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.2, - "output": 6 + "input": 0.56, + "output": 6.67 } }, - "qwen/qwen3.5-plus": { - "id": "qwen/qwen3.5-plus", - "name": "Qwen3.5 Plus", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/meta/llama-3.2-3b-instruct": { + "id": "workers-ai/@cf/meta/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-03-20", - "last_updated": "2026-03-20", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.8, - "output": 4.8 + "input": 0.051, + "output": 0.34 } }, - "google/gemini-3.1-flash-lite-preview": { - "id": "google/gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "attachment": true, + "workers-ai/@cf/facebook/bart-large-cnn": { + "id": "workers-ai/@cf/facebook/bart-large-cnn", + "name": "BART Large CNN", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "bart", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2025-03-20", - "last_updated": "2025-03-20", + "release_date": "2025-04-09", + "last_updated": "2025-04-09", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 65530 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 1.5 + "input": 0, + "output": 0 } }, - "google/gemini-3.1-pro-preview": { - "id": "google/gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "attachment": true, - "reasoning": true, - "tool_call": true, + "workers-ai/@cf/aisingapore/gemma-sea-lion-v4-27b-it": { + "id": "workers-ai/@cf/aisingapore/gemma-sea-lion-v4-27b-it", + "name": "Gemma SEA-LION v4 27B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2026-02-19", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image", "pdf", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "cache_write": 4.5 + "input": 0.35, + "output": 0.56 } }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text", "image", "pdf", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "output": 64000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 1 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "attachment": true, - "reasoning": true, - "tool_call": true, + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2021-09-01", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", "modalities": { - "input": ["pdf", "image", "text", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "output": 64000 + "context": 16385, + "output": 4096 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31, - "cache_write": 4.5 + "input": 0.5, + "output": 1.5, + "cache_read": 1.25 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["pdf", "image", "text", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.07, - "cache_write": 1 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "google/gemini-2.5-flash-lite": { - "id": "google/gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", + "openai/gpt-4": { + "id": "openai/gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["pdf", "image", "text", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048000, - "output": 64000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.03, - "cache_write": 1 + "input": 30, + "output": 60 } }, - "sapiens-ai/agnes-1.5-lite": { - "id": "sapiens-ai/agnes-1.5-lite", - "name": "Agnes 1.5 Lite", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-26", - "last_updated": "2026-03-26", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.12, - "output": 0.6 + "input": 1.1, + "output": 4.4, + "cache_read": 0.28 } }, - "sapiens-ai/agnes-1.5-pro": { - "id": "sapiens-ai/agnes-1.5-pro", - "name": "Agnes 1.5 Pro", - "attachment": false, + "openai/o3-pro": { + "id": "openai/o3-pro", + "name": "o3-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-21", - "last_updated": "2026-03-21", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-06-10", + "last_updated": "2025-06-10", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.16, - "output": 0.8 + "input": 20, + "output": 80 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": false, - "knowledge": "2025-01-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.58, - "output": 3.02, - "cache_read": 0.1 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "moonshotai/kimi-k2-thinking-turbo": { - "id": "moonshotai/kimi-k2-thinking-turbo", - "name": "Kimi K2 Thinking Turbo", + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 64000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 1.15, - "output": 8, - "cache_read": 0.15 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 0905", - "attachment": false, - "reasoning": false, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-04", - "last_updated": "2025-09-04", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": false, - "knowledge": "2025-01-01", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262140, - "output": 262140 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "ai-gateway-provider" }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "attachment": false, + "openai/o1": { + "id": "openai/o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 64000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Claude Opus 4.1", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["image", "text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "npm": "ai-gateway-provider" }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "anthropic/claude-3.7-sonnet": { - "id": "anthropic/claude-3.7-sonnet", - "name": "Claude 3.7 Sonnet", + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 4096 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 10, + "output": 30 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Claude Opus 4.6", + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.15, + "output": 0.6, + "cache_read": 0.08 } }, - "anthropic/claude-opus-4.7": { - "id": "anthropic/claude-opus-4.7", - "name": "Claude Opus 4.7", + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "structured_output": true, "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, + "input": 272000, "output": 128000 }, "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "npm": "ai-gateway-provider" }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4", + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["image", "text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 400000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "anthropic/claude-sonnet-4.5": { - "id": "anthropic/claude-sonnet-4.5", - "name": "Claude Sonnet 4.5", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } }, - "anthropic/claude-opus-4.5": { - "id": "anthropic/claude-opus-4.5", - "name": "Claude Opus 4.5", + "anthropic/claude-3.5-haiku": { + "id": "anthropic/claude-3.5-haiku", + "name": "Claude Haiku 3.5 (latest)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["pdf", "image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "output": 8192 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude Opus 4", + "anthropic/claude-3.5-sonnet": { + "id": "anthropic/claude-3.5-sonnet", + "name": "Claude Sonnet 3.5 v2", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2024-04-30", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["image", "text", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 32000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" + "output": 8192 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "anthropic/claude-3.5-haiku": { - "id": "anthropic/claude-3.5-haiku", - "name": "Claude 3.5 Haiku", + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2024-11-04", - "last_updated": "2024-11-04", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, @@ -37016,29 +78593,33 @@ "context": 200000, "output": 64000 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" - }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "anthropic/claude-haiku-4.5": { - "id": "anthropic/claude-haiku-4.5", - "name": "Claude Haiku 4.5", + "anthropic/claude-opus-4-5": { + "id": "anthropic/claude-opus-4-5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, @@ -37046,40 +78627,40 @@ "context": 200000, "output": 64000 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" - }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Claude Sonnet 4.6", + "anthropic/claude-sonnet-4-5": { + "id": "anthropic/claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-18", - "last_updated": "2026-02-18", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 200000, "output": 64000 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://zenmux.ai/api/anthropic/v1" - }, "cost": { "input": 3, "output": 15, @@ -37087,1826 +78668,2260 @@ "cache_write": 3.75 } }, - "deepseek/deepseek-v4-flash": { - "id": "deepseek/deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", - "attachment": false, + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepseek/deepseek-v4-pro": { - "id": "deepseek/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", - "attachment": false, + "anthropic/claude-sonnet-5": { + "id": "anthropic/claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "interleaved": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 128000 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "tencent/hy3-preview": { - "id": "tencent/hy3-preview", - "name": "Hy3 preview", - "family": "Hy", - "attachment": false, - "reasoning": true, + "anthropic/claude-3-sonnet": { + "id": "anthropic/claude-3-sonnet", + "name": "Claude Sonnet 3", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "knowledge": "2023-08-31", + "release_date": "2024-03-04", + "last_updated": "2024-03-04", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 200000, + "output": 4096 }, "cost": { - "input": 0.172, - "output": 0.572, - "cache_read": 0.058, - "cache_write": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 0.3 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "anthropic/claude-opus-4-8": { + "id": "anthropic/claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, + "context": 1000000, "output": 128000 }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 12.5, - "output": 75, - "cache_read": 1.25 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } - }, "cost": { "input": 5, - "output": 30, + "output": 25, "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "cache_write": 6.25 } }, - "openai/gpt-5.5-pro": { - "id": "openai/gpt-5.5-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", + "anthropic/claude-3-opus": { + "id": "anthropic/claude-3-opus", + "name": "Claude Opus 3", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "temperature": true, + "knowledge": "2023-08-31", + "release_date": "2024-02-29", + "last_updated": "2024-02-29", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 200000, + "output": 4096 }, "cost": { - "input": 30, - "output": 180, - "context_over_200k": { - "input": 60, - "output": 270 - }, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "xiaomi/mimo-v2.5-pro": { - "id": "xiaomi/mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", - "family": "mimo", - "attachment": false, - "reasoning": true, + "anthropic/claude-3-5-haiku": { + "id": "anthropic/claude-3-5-haiku", + "name": "Claude Haiku 3.5 (latest)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 200000, + "output": 8192 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "xiaomi/mimo-v2-omni": { - "id": "xiaomi/mimo-v2-omni", - "name": "MiMo V2 Omni", - "family": "mimo", + "anthropic/claude-opus-4-1": { + "id": "anthropic/claude-opus-4-1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 265000, - "output": 265000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "xiaomi/mimo-v2.5": { - "id": "xiaomi/mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", + "anthropic/claude-fable-5": { + "id": "anthropic/claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - }, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "xiaomi/mimo-v2-pro": { - "id": "xiaomi/mimo-v2-pro", - "name": "MiMo V2 Pro", - "family": "mimo", - "attachment": false, + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 256000 + "context": 200000, + "output": 64000 }, "cost": { "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "MiMo-V2-Flash", - "family": "mimo", - "attachment": false, + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } - } - } - }, - "upstage": { - "id": "upstage", - "env": ["UPSTAGE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.upstage.ai/v1/solar", - "name": "Upstage", - "doc": "https://developers.upstage.ai/docs/apis/chat", - "models": { - "solar-pro2": { - "id": "solar-pro2", - "name": "solar-pro2", - "family": "solar-pro", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + }, + "anthropic/claude-3-haiku": { + "id": "anthropic/claude-3-haiku", + "name": "Claude Haiku 3", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2023-08-31", + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 8192 + "context": 200000, + "output": 4096 }, "cost": { "input": 0.25, - "output": 0.25 + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 } }, - "solar-mini": { - "id": "solar-mini", - "name": "solar-mini", - "family": "solar-mini", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-06-12", - "last_updated": "2025-04-22", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 4096 + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "ai-gateway-provider" }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "solar-pro3": { - "id": "solar-pro3", - "name": "solar-pro3", - "family": "solar-pro", - "attachment": false, + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2026-01", - "last_updated": "2026-01", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 0.25 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } } } }, - "novita-ai": { - "id": "novita-ai", - "env": ["NOVITA_API_KEY"], + "moonshotai-cn": { + "id": "moonshotai-cn", + "env": ["MOONSHOT_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.novita.ai/openai", - "name": "NovitaAI", - "doc": "https://novita.ai/docs/guides/introduction", + "api": "https://api.moonshot.cn/v1", + "name": "Moonshot AI (China)", + "doc": "https://platform.moonshot.cn/docs/api/chat", "models": { - "deepseek/deepseek-r1-turbo": { - "id": "deepseek/deepseek-r1-turbo", - "name": "DeepSeek R1 (Turbo)\t", - "attachment": false, + "kimi-k2.7-code-highspeed": { + "id": "kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code HighSpeed", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 64000, - "output": 16000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.7, - "output": 2.5 + "input": 1.9, + "output": 8, + "cache_read": 0.38 } }, - "deepseek/deepseek-v3-0324": { - "id": "deepseek/deepseek-v3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 163840 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27, - "output": 1.12, - "cache_read": 0.135 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "deepseek/deepseek-ocr-2": { - "id": "deepseek/deepseek-ocr-2", - "name": "deepseek/deepseek-ocr-2", - "attachment": true, - "reasoning": false, - "tool_call": false, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 8192, - "output": 8192 + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.03, - "output": 0.03 - } - }, - "deepseek/deepseek-ocr": { - "id": "deepseek/deepseek-ocr", - "name": "DeepSeek-OCR", - "attachment": true, - "reasoning": false, - "tool_call": false, "structured_output": true, - "temperature": true, - "release_date": "2025-10-24", - "last_updated": "2025-10-24", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.03, - "output": 0.03 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "deepseek/deepseek-r1-distill-llama-70b": { - "id": "deepseek/deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill LLama 70B", - "family": "deepseek-thinking", + "kimi-k2-turbo-preview": { + "id": "kimi-k2-turbo-preview", + "name": "Kimi K2 Turbo", + "description": "Fast Kimi model for responsive chat, coding help, and agent loops", + "family": "kimi-k2", "attachment": false, - "reasoning": true, - "tool_call": false, - "structured_output": true, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.8, - "output": 0.8 + "input": 2.4, + "output": 10, + "cache_read": 0.6 } }, - "deepseek/deepseek-prover-v2-671b": { - "id": "deepseek/deepseek-prover-v2-671b", - "name": "Deepseek Prover V2 671B", + "kimi-k2-0711-preview": { + "id": "kimi-k2-0711-preview", + "name": "Kimi K2 0711", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-04-30", + "knowledge": "2024-10", + "release_date": "2025-07-14", + "last_updated": "2025-07-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 160000, - "output": 160000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.7, - "output": 2.5 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "deepseek/deepseek-r1-0528-qwen3-8b": { - "id": "deepseek/deepseek-r1-0528-qwen3-8b", - "name": "DeepSeek R1 0528 Qwen3 8B", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-05-29", - "last_updated": "2025-05-29", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.06, - "output": 0.09 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "deepseek/deepseek-r1-distill-qwen-32b": { - "id": "deepseek/deepseek-r1-distill-qwen-32b", - "name": "DeepSeek R1 Distill Qwen 32B", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 64000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "deepseek/deepseek-v3.2-exp": { - "id": "deepseek/deepseek-v3.2-exp", - "name": "Deepseek V3.2 Exp", + "kimi-k2-thinking-turbo": { + "id": "kimi-k2-thinking-turbo", + "name": "Kimi K2 Thinking Turbo", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27, - "output": 0.41 + "input": 1.15, + "output": 8, + "cache_read": 0.15 } }, - "deepseek/deepseek-v3.1": { - "id": "deepseek/deepseek-v3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "kimi-k2-0905-preview": { + "id": "kimi-k2-0905-preview", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27, - "output": 1, - "cache_read": 0.135 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } - }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "Deepseek V3.2", - "family": "deepseek", + } + } + }, + "morph": { + "id": "morph", + "env": ["MORPH_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.morphllm.com/v1", + "name": "Morph", + "doc": "https://docs.morphllm.com/api-reference/introduction", + "models": { + "morph-v3-fast": { + "id": "morph-v3-fast", + "name": "Morph v3 Fast", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "morph", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 16000, + "output": 16000 }, "cost": { - "input": 0.269, - "output": 0.4, - "cache_read": 0.1345 + "input": 0.8, + "output": 1.2 } }, - "deepseek/deepseek-v3-turbo": { - "id": "deepseek/deepseek-v3-turbo", - "name": "DeepSeek V3 (Turbo)\t", + "morph-v3-large": { + "id": "morph-v3-large", + "name": "Morph v3 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "morph", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "tool_call": false, + "temperature": false, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 16000 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.4, - "output": 1.3 + "input": 0.9, + "output": 1.9 } }, - "deepseek/deepseek-r1-distill-qwen-14b": { - "id": "deepseek/deepseek-r1-distill-qwen-14b", - "name": "DeepSeek R1 Distill Qwen 14B", - "family": "deepseek-thinking", + "auto": { + "id": "auto", + "name": "Auto", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "temperature": false, + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 32000, + "output": 32000 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.85, + "output": 1.55 } - }, - "deepseek/deepseek-r1-0528": { - "id": "deepseek/deepseek-r1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek-thinking", - "attachment": false, + } + } + }, + "sakana": { + "id": "sakana", + "env": ["SAKANA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.sakana.ai/v1", + "name": "Sakana AI", + "doc": "https://console.sakana.ai/models", + "models": { + "fugu-ultra-20260615": { + "id": "fugu-ultra-20260615", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "fugu", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "temperature": false, + "release_date": "2026-06-15", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 + "context": 1000000, + "output": 1000000 + }, + "provider": { + "shape": "responses" }, "cost": { - "input": 0.7, - "output": 2.5, - "cache_read": 0.35 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } }, - "deepseek/deepseek-v3.1-terminus": { - "id": "deepseek/deepseek-v3.1-terminus", - "name": "Deepseek V3.1 Terminus", - "family": "deepseek", - "attachment": false, + "fugu": { + "id": "fugu", + "name": "Fugu", + "description": "Multi-agent model for routing expert agents across complex analytical tasks", + "family": "fugu", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "temperature": false, + "release_date": "2026-06-15", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "output": 1000000 }, - "cost": { - "input": 0.27, - "output": 1, - "cache_read": 0.135 + "provider": { + "shape": "responses" } }, - "inclusionai/ling-2.6-1t": { - "id": "inclusionai/ling-2.6-1t", - "name": "Ling-2.6-1T", - "family": "ling", - "attachment": false, - "reasoning": false, + "fugu-ultra": { + "id": "fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "fugu", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "temperature": false, + "release_date": "2026-06-15", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 1000000 + }, + "provider": { + "shape": "responses" }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } - }, - "inclusionai/ling-2.6-flash": { - "id": "inclusionai/ling-2.6-flash", - "name": "Ling-2.6-flash", - "family": "ling", - "attachment": false, + } + } + }, + "deepinfra": { + "id": "deepinfra", + "env": ["DEEPINFRA_API_KEY"], + "npm": "@ai-sdk/deepinfra", + "name": "Deep Infra", + "doc": "https://deepinfra.com/models", + "models": { + "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { + "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", + "name": "Llama 4 Maverick 17B FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": true, - "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.02 + "input": 0.2, + "output": 0.8 } }, - "paddlepaddle/paddleocr-vl": { - "id": "paddlepaddle/paddleocr-vl", - "name": "PaddleOCR-VL", + "meta-llama/Llama-4-Scout-17B-16E-Instruct": { + "id": "meta-llama/Llama-4-Scout-17B-16E-Instruct", + "name": "Llama 4 Scout 17B", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-10-22", - "last_updated": "2025-10-22", + "tool_call": true, + "structured_output": true, + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, + "context": 327680, "output": 16384 }, "cost": { - "input": 0.02, - "output": 0.02 + "input": 0.1, + "output": 0.3 } }, - "nousresearch/hermes-2-pro-llama-3-8b": { - "id": "nousresearch/hermes-2-pro-llama-3-8b", - "name": "Hermes 2 Pro Llama 3 8B", + "meta-llama/Llama-3.3-70B-Instruct-Turbo": { + "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo", + "name": "Llama 3.3 70B Turbo", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2024-06-27", - "last_updated": "2024-06-27", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.14 + "input": 0.1, + "output": 0.32 } }, - "zai-org/glm-4.7": { - "id": "zai-org/glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-04", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11 + "input": 0.75, + "output": 3.5, + "cache_read": 0.15 } }, - "zai-org/glm-5": { - "id": "zai-org/glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-12", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202800, - "output": 131072 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.45, + "output": 2.25, + "cache_read": 0.07 } }, - "zai-org/glm-5.1": { - "id": "zai-org/glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, - "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.74, + "output": 3.5, + "cache_read": 0.15 } }, - "zai-org/glm-4.5": { - "id": "zai-org/glm-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11 + "input": 0.13, + "output": 0.38 } }, - "zai-org/glm-4.5-air": { - "id": "zai-org/glm-4.5-air", - "name": "GLM 4.5 Air", - "family": "glm-air", - "attachment": false, + "google/gemma-4-26B-A4B-it": { + "id": "google/gemma-4-26B-A4B-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-10-13", - "last_updated": "2025-10-13", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.13, - "output": 0.85 + "input": 0.07, + "output": 0.34 } }, - "zai-org/glm-4.5v": { - "id": "zai-org/glm-4.5v", - "name": "GLM 4.5V", - "family": "glmv", + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "video", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 16384 + "context": 262144, + "output": 81920 }, "cost": { - "input": 0.6, - "output": 1.8, - "cache_read": 0.11 + "input": 0.15, + "output": 0.95 } }, - "zai-org/glm-4.6": { - "id": "zai-org/glm-4.6", - "name": "GLM 4.6", - "family": "glm", + "Qwen/Qwen3.7-Max": { + "id": "Qwen/Qwen3.7-Max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.55, - "output": 2.2, - "cache_read": 0.11 + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "tiers": [ + { + "input": 5, + "output": 15, + "cache_read": 1, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 6.25, + "output": 18.5, + "cache_read": 1.25, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "zai-org/glm-4.6v": { - "id": "zai-org/glm-4.6v", - "name": "GLM 4.6V", - "family": "glmv", + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.32, + "output": 3.2 + } + }, + "Qwen/Qwen3-Next-80B-A3B-Instruct": { + "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", + "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "video", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 262144, "output": 32768 }, "cost": { - "input": 0.3, - "output": 0.9, - "cache_read": 0.055 + "input": 0.09, + "output": 1.1 } }, - "zai-org/glm-4.7-flash": { - "id": "zai-org/glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm", + "Qwen/Qwen3-Max": { + "id": "Qwen/Qwen3-Max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.4, - "cache_read": 0.01 + "input": 1.2, + "output": 6, + "cache_read": 0.24, + "tiers": [ + { + "input": 2.4, + "output": 12, + "cache_read": 0.48, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 3, + "output": 15, + "cache_read": 0.6, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "zai-org/autoglm-phone-9b-multilingual": { - "id": "zai-org/autoglm-phone-9b-multilingual", - "name": "AutoGLM-Phone-9B-Multilingual", + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen 3.5 397B A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-12-10", - "last_updated": "2025-12-10", + "knowledge": "2025-01", + "release_date": "2026-02-01", + "last_updated": "2026-04-20", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 262144, + "output": 81920 }, "cost": { - "input": 0.035, - "output": 0.138 + "input": 0.45, + "output": 3, + "cache_read": 0.22 } }, - "mistralai/mistral-nemo": { - "id": "mistralai/mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, - "tool_call": false, + "Qwen/Qwen3.5-122B-A10B": { + "id": "Qwen/Qwen3.5-122B-A10B", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-07-30", - "last_updated": "2024-07-30", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 60288, - "output": 16000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.04, - "output": 0.17 + "input": 0.29, + "output": 2.4 } }, - "baichuan/baichuan-m2-32b": { - "id": "baichuan/baichuan-m2-32b", - "name": "baichuan-m2-32b", - "family": "baichuan", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "Qwen/Qwen3.5-27B": { + "id": "Qwen/Qwen3.5-27B", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-08-13", - "last_updated": "2025-08-13", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.07 + "input": 0.26, + "output": 2.6 } }, - "meta-llama/llama-4-scout-17b-16e-instruct": { - "id": "meta-llama/llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout Instruct", + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-06", - "last_updated": "2025-04-06", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.18, - "output": 0.59 + "input": 0.1, + "output": 0.15 } }, - "meta-llama/llama-3.3-70b-instruct": { - "id": "meta-llama/llama-3.3-70b-instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-07", - "last_updated": "2024-12-07", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 120000 + "context": 40960, + "output": 16384 }, "cost": { - "input": 0.135, - "output": 0.4 + "input": 0.08, + "output": 0.28 } }, - "meta-llama/llama-3.2-3b-instruct": { - "id": "meta-llama/llama-3.2-3b-instruct", - "name": "Llama 3.2 3B Instruct", - "family": "llama", + "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo", + "name": "Qwen3 Coder 480B A35B Instruct Turbo", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32000 + "context": 262144, + "output": 66536 }, "cost": { - "input": 0.03, - "output": 0.05 + "input": 0.3, + "output": 1, + "cache_read": 0.1 } }, - "meta-llama/llama-3-8b-instruct": { - "id": "meta-llama/llama-3-8b-instruct", - "name": "Llama 3 8B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "Qwen/Qwen3.5-35B-A3B": { + "id": "Qwen/Qwen3.5-35B-A3B", + "name": "Qwen 3.5 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-04-25", - "last_updated": "2024-04-25", + "knowledge": "2025-01", + "release_date": "2026-02-01", + "last_updated": "2026-04-20", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 81920 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 0.14, + "output": 1, + "cache_read": 0.05 } }, - "meta-llama/llama-3.1-8b-instruct": { - "id": "meta-llama/llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-07-24", - "last_updated": "2024-07-24", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, + "context": 131072, "output": 16384 }, "cost": { - "input": 0.02, - "output": 0.05 + "input": 0.037, + "output": 0.17 } }, - "meta-llama/llama-3-70b-instruct": { - "id": "meta-llama/llama-3-70b-instruct", - "name": "Llama3 70B Instruct", - "family": "llama", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-04-25", - "last_updated": "2024-04-25", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.51, - "output": 0.74 + "input": 0.03, + "output": 0.14 } }, - "meta-llama/llama-4-maverick-17b-128e-instruct-fp8": { - "id": "meta-llama/llama-4-maverick-17b-128e-instruct-fp8", - "name": "Llama 4 Maverick Instruct", + "XiaomiMiMo/MiMo-V2.5": { + "id": "XiaomiMiMo/MiMo-V2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-04-06", - "last_updated": "2025-04-06", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 8192 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.27, - "output": 0.85 + "input": 0.4, + "output": 2, + "cache_read": 0.08 } }, - "gryphe/mythomax-l2-13b": { - "id": "gryphe/mythomax-l2-13b", - "name": "Mythomax L2 13B", - "attachment": false, - "reasoning": false, - "tool_call": false, + "XiaomiMiMo/MiMo-V2.5-Pro": { + "id": "XiaomiMiMo/MiMo-V2.5-Pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-04-25", - "last_updated": "2024-04-25", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 4096, - "output": 3200 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 0.09, - "output": 0.09 + "input": 1, + "output": 3, + "cache_read": 0.2 } }, - "sao10k/l31-70b-euryale-v2.2": { - "id": "sao10k/l31-70b-euryale-v2.2", - "name": "L31 70B Euryale V2.2", - "attachment": false, - "reasoning": false, + "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning": { + "id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning", + "name": "Nemotron 3 Nano Omni 30B A3B Reasoning", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.48, - "output": 1.48 + "input": 0.2, + "output": 0.8 } }, - "sao10k/l3-70b-euryale-v2.1": { - "id": "sao10k/l3-70b-euryale-v2.1", - "name": "L3 70B Euryale V2.1\t", + "nvidia/Nemotron-3-Nano-30B-A3B": { + "id": "nvidia/Nemotron-3-Nano-30B-A3B", + "name": "Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-06-18", - "last_updated": "2024-06-18", + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.48, - "output": 1.48 + "input": 0.05, + "output": 0.2 } }, - "sao10k/l3-8b-lunaris": { - "id": "sao10k/l3-8b-lunaris", - "name": "Sao10k L3 8B Lunaris\t", + "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": { + "id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5", + "name": "Llama 3.3 Nemotron Super 49B v1.5", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-11-28", - "last_updated": "2024-11-28", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 0.4, + "output": 0.4 } }, - "microsoft/wizardlm-2-8x22b": { - "id": "microsoft/wizardlm-2-8x22b", - "name": "Wizardlm 2 8x22B", + "zai-org/GLM-4.7-Flash": { + "id": "zai-org/GLM-4.7-Flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-04-24", - "last_updated": "2024-04-24", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65535, - "output": 8000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.62, - "output": 0.62 + "input": 0.06, + "output": 0.4, + "cache_read": 0.01 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "OpenAI: GPT OSS 20B", - "attachment": true, + "zai-org/GLM-4.6": { + "id": "zai-org/GLM-4.6", + "name": "GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.04, - "output": 0.15 + "input": 0.5, + "output": 2, + "cache_read": 0.1 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "OpenAI GPT OSS 120B", - "attachment": true, + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "knowledge": "2025-12", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.25 + "input": 0.6, + "output": 2.08, + "cache_read": 0.12 } }, - "minimaxai/minimax-m1-80k": { - "id": "minimaxai/minimax-m1-80k", - "name": "MiniMax M1", - "family": "minimax", + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 40000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 0.4, + "output": 1.75, + "cache_read": 0.08 } }, - "Sao10K/L3-8B-Stheno-v3.2": { - "id": "Sao10K/L3-8B-Stheno-v3.2", - "name": "L3 8B Stheno V3.2", - "family": "llama", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-11-29", - "last_updated": "2024-11-29", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 32000 + "context": 1048576, + "output": 32768 }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 0.93, + "output": 3, + "cache_read": 0.18 } }, - "xiaomimimo/mimo-v2-flash": { - "id": "xiaomimimo/mimo-v2-flash", - "name": "XiaomiMiMo/MiMo-V2-Flash", - "family": "mimo", + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "knowledge": "2025-04", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.3 + "input": 1.05, + "output": 3.5, + "cache_read": 0.205 } }, - "baidu/ernie-4.5-vl-28b-a3b-thinking": { - "id": "baidu/ernie-4.5-vl-28b-a3b-thinking", - "name": "ERNIE-4.5-VL-28B-A3B-Thinking", - "attachment": true, + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-11-26", - "last_updated": "2025-11-26", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 163840, + "output": 64000 }, "cost": { - "input": 0.39, - "output": 0.39 + "input": 0.5, + "output": 2.15, + "cache_read": 0.35 } }, - "baidu/ernie-4.5-vl-424b-a47b": { - "id": "baidu/ernie-4.5-vl-424b-a47b", - "name": "ERNIE 4.5 VL 424B A47B", - "attachment": true, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 123000, - "output": 16000 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 0.42, - "output": 1.25 + "input": 0.09, + "output": 0.18, + "cache_read": 0.018 } }, - "baidu/ernie-4.5-21B-a3b": { - "id": "baidu/ernie-4.5-21B-a3b", - "name": "ERNIE 4.5 21B A3B", - "family": "ernie", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 120000, - "output": 8000 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 1.3, + "output": 2.6, + "cache_read": 0.1 } }, - "baidu/ernie-4.5-300b-a47b-paddle": { - "id": "baidu/ernie-4.5-300b-a47b-paddle", - "name": "ERNIE 4.5 300B A47B", + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "knowledge": "2024-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 123000, - "output": 12000 + "context": 163840, + "output": 64000 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 0.26, + "output": 0.38, + "cache_read": 0.13 } }, - "baidu/ernie-4.5-21B-a3b-thinking": { - "id": "baidu/ernie-4.5-21B-a3b-thinking", - "name": "ERNIE-4.5-21B-A3B-Thinking", - "family": "ernie", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "knowledge": "2025-06", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 196608, + "output": 131072 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.15, + "output": 1.15, + "cache_read": 0.03 } }, - "baidu/ernie-4.5-vl-28b-a3b": { - "id": "baidu/ernie-4.5-vl-28b-a3b", - "name": "ERNIE 4.5 VL 28B A3B", + "MiniMaxAI/MiniMax-M3": { + "id": "MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 30000, - "output": 8000 + "context": 524288, + "output": 128000 }, "cost": { - "input": 1.4, - "output": 5.6 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "MiniMax M2.7", - "family": "minimax-m2.7", + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, "release_date": "2026-03-18", "last_updated": "2026-03-18", @@ -38916,726 +80931,993 @@ }, "open_weights": true, "limit": { - "context": 204800, + "context": 196608, "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0.25, + "output": 1, + "cache_read": 0.05 } - }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "MiniMax-M2", - "family": "minimax", - "attachment": false, + } + } + }, + "google-vertex-anthropic": { + "id": "google-vertex-anthropic", + "env": ["GOOGLE_VERTEX_PROJECT", "GOOGLE_VERTEX_LOCATION", "GOOGLE_APPLICATION_CREDENTIALS"], + "npm": "@ai-sdk/google-vertex/anthropic", + "name": "Vertex (Anthropic)", + "doc": "https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude", + "models": { + "claude-haiku-4-5@20251001": { + "id": "claude-haiku-4-5@20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "Minimax M2.1", - "family": "minimax", - "attachment": false, - "reasoning": false, + "claude-opus-4@20250514": { + "id": "claude-opus-4@20250514", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 32000 }, + "status": "deprecated", "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "claude-opus-4-1@20250805": { + "id": "claude-opus-4-1@20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131100 + "context": 200000, + "output": 32000 }, + "status": "deprecated", "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "minimax/minimax-m2.5-highspeed": { - "id": "minimax/minimax-m2.5-highspeed", - "name": "MiniMax M2.5 Highspeed", - "family": "minimax-m2.5", - "attachment": false, + "claude-opus-4-5@20251101": { + "id": "claude-opus-4-5@20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-05", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131100 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.03 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen/qwen2.5-7b-instruct": { - "id": "qwen/qwen2.5-7b-instruct", - "name": "Qwen2.5 7B Instruct", - "attachment": false, + "claude-3-5-haiku@20241022": { + "id": "claude-3-5-haiku@20241022", + "name": "Claude Haiku 3.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 32000 + "context": 200000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.07, - "output": 0.07 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "qwen/qwen3.5-122b-a10b": { - "id": "qwen/qwen3.5-122b-a10b", - "name": "Qwen3.5-122B-A10B", - "family": "qwen", + "claude-sonnet-4@20250514": { + "id": "claude-sonnet-4@20250514", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 200000, + "output": 64000 }, + "status": "deprecated", "cost": { - "input": 0.4, - "output": 3.2 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "qwen/qwen3.5-27b": { - "id": "qwen/qwen3.5-27b", - "name": "Qwen3.5-27B", - "family": "qwen", + "claude-opus-4-7@default": { + "id": "claude-opus-4-7@default", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 2.4 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "qwen/qwen3-235b-a22b-instruct-2507": { - "id": "qwen/qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-5@20250929": { + "id": "claude-sonnet-4-5@20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.09, - "output": 0.58 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "qwen/qwen3-omni-30b-a3b-instruct": { - "id": "qwen/qwen3-omni-30b-a3b-instruct", - "name": "Qwen3 Omni 30B A3B Instruct", - "family": "qwen", + "claude-sonnet-5@default": { + "id": "claude-sonnet-5@default", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "video", "audio", "image"], - "output": ["text", "audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 0.97, - "input_audio": 2.2, - "output_audio": 1.788 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen3.5-397B-A17B", - "family": "qwen", + "claude-opus-4-6@default": { + "id": "claude-opus-4-6@default", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "qwen/qwen2.5-vl-72b-instruct": { - "id": "qwen/qwen2.5-vl-72b-instruct", - "name": "Qwen2.5 VL 72B Instruct", - "family": "qwen", + "claude-opus-4-8@default": { + "id": "claude-opus-4-8@default", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.8, - "output": 0.8 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "qwen/qwen3-vl-235b-a22b-thinking": { - "id": "qwen/qwen3-vl-235b-a22b-thinking", - "name": "Qwen3 VL 235B A22B Thinking", + "claude-sonnet-4-6@default": { + "id": "claude-sonnet-4-6@default", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.98, - "output": 3.95 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } - }, - "qwen/qwen3-vl-30b-a3b-thinking": { - "id": "qwen/qwen3-vl-30b-a3b-thinking", - "name": "qwen/qwen3-vl-30b-a3b-thinking", + } + } + }, + "v0": { + "id": "v0", + "env": ["V0_API_KEY"], + "npm": "@ai-sdk/vercel", + "name": "v0", + "doc": "https://sdk.vercel.ai/providers/ai-sdk-providers/vercel", + "models": { + "v0-1.0-md": { + "id": "v0-1.0-md", + "name": "v0-1.0-md", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "v0", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2025-10-11", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 1 + "input": 3, + "output": 15 } }, - "qwen/qwen3-omni-30b-a3b-thinking": { - "id": "qwen/qwen3-omni-30b-a3b-thinking", - "name": "Qwen3 Omni 30B A3B Thinking", + "v0-1.5-lg": { + "id": "v0-1.5-lg", + "name": "v0-1.5-lg", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "v0", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "release_date": "2025-06-09", + "last_updated": "2025-06-09", "modalities": { - "input": ["text", "audio", "video", "image"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 512000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 0.97, - "input_audio": 2.2, - "output_audio": 1.788 + "input": 15, + "output": 75 } }, - "qwen/qwen3-vl-8b-instruct": { - "id": "qwen/qwen3-vl-8b-instruct", - "name": "qwen/qwen3-vl-8b-instruct", + "v0-1.5-md": { + "id": "v0-1.5-md", + "name": "v0-1.5-md", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "v0", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-17", - "last_updated": "2025-10-17", + "release_date": "2025-06-09", + "last_updated": "2025-06-09", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.08, - "output": 0.5 + "input": 3, + "output": 15 } - }, - "qwen/qwen3-max": { - "id": "qwen/qwen3-max", - "name": "Qwen3 Max", - "family": "qwen", - "attachment": false, - "reasoning": false, + } + } + }, + "azure": { + "id": "azure", + "env": ["AZURE_RESOURCE_NAME", "AZURE_API_KEY"], + "npm": "@ai-sdk/azure", + "name": "Azure", + "doc": "https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models", + "models": { + "codex-mini": { + "id": "codex-mini", + "name": "Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-05-16", + "last_updated": "2025-05-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 2.11, - "output": 8.45 + "input": 1.5, + "output": 6, + "cache_read": 0.375 } }, - "qwen/qwen3-32b-fp8": { - "id": "qwen/qwen3-32b-fp8", - "name": "Qwen3 32B", + "phi-3.5-moe-instruct": { + "id": "phi-3.5-moe-instruct", + "name": "Phi-3.5-MoE-instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 20000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.45 + "input": 0.16, + "output": 0.64 } }, - "qwen/qwen3-4b-fp8": { - "id": "qwen/qwen3-4b-fp8", - "name": "Qwen3 4B", + "gpt-3.5-turbo-instruct": { + "id": "gpt-3.5-turbo-instruct", + "name": "GPT-3.5 Turbo Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "knowledge": "2021-08", + "release_date": "2023-09-21", + "last_updated": "2023-09-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 20000 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0.03, - "output": 0.03 + "input": 1.5, + "output": 2 } }, - "qwen/qwen3-235b-a22b-thinking-2507": { - "id": "qwen/qwen3-235b-a22b-thinking-2507", - "name": "Qwen3 235B A22b Thinking 2507", - "family": "qwen", + "deepseek-r1-0528": { + "id": "deepseek-r1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.3, - "output": 3 + "input": 1.35, + "output": 5.4 } }, - "qwen/qwen3-next-80b-a3b-thinking": { - "id": "qwen/qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek-V4-Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "structured_output": true, "temperature": true, - "release_date": "2025-09-10", - "last_updated": "2025-09-10", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "output": 384000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 0.15, - "output": 1.5 + "input": 0.19, + "output": 0.51 } }, - "qwen/qwen-mt-plus": { - "id": "qwen/qwen-mt-plus", - "name": "Qwen MT Plus", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-09-03", - "last_updated": "2025-09-03", + "gpt-5.2-chat": { + "id": "gpt-5.2-chat", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 0.75 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "qwen/qwen3-next-80b-a3b-instruct": { - "id": "qwen/qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "attachment": false, - "reasoning": false, + "o3": { + "id": "o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-09-10", - "last_updated": "2025-09-10", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.15, - "output": 1.5 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "qwen/qwen3-30b-a3b-fp8": { - "id": "qwen/qwen3-30b-a3b-fp8", - "name": "Qwen3 30B A3B", + "deepseek-v3-0324": { + "id": "deepseek-v3-0324", + "name": "DeepSeek-V3-0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "knowledge": "2024-07", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 20000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.09, - "output": 0.45 + "input": 1.14, + "output": 4.56 } }, - "qwen/qwen3-coder-next": { - "id": "qwen/qwen3-coder-next", - "name": "Qwen3 Coder Next", - "family": "qwen", + "phi-3-small-128k-instruct": { + "id": "phi-3-small-128k-instruct", + "name": "Phi-3-small-instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2026-02-03", - "last_updated": "2026-02-03", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 0.15, + "output": 0.6 } }, - "qwen/qwen3-coder-480b-a35b-instruct": { - "id": "qwen/qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", + "meta-llama-3-8b-instruct": { + "id": "meta-llama-3-8b-instruct", + "name": "Meta-Llama-3-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 8192, + "output": 2048 }, "cost": { "input": 0.3, - "output": 1.3 + "output": 0.61 } }, - "qwen/qwen3-vl-30b-a3b-instruct": { - "id": "qwen/qwen3-vl-30b-a3b-instruct", - "name": "qwen/qwen3-vl-30b-a3b-instruct", + "mistral-small-2503": { + "id": "mistral-small-2503", + "name": "Mistral Small 3.1", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2025-10-11", + "knowledge": "2024-09", + "release_date": "2025-03-01", + "last_updated": "2025-03-01", "modalities": { - "input": ["text", "video", "image"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 128000, "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.7 + "input": 0.1, + "output": 0.3 } }, - "qwen/qwen3-coder-30b-a3b-instruct": { - "id": "qwen/qwen3-coder-30b-a3b-instruct", - "name": "Qwen3 Coder 30b A3B Instruct", + "text-embedding-3-large": { + "id": "text-embedding-3-large", + "name": "text-embedding-3-large", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-09", - "last_updated": "2025-10-09", + "tool_call": false, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 160000, - "output": 32768 + "context": 8191, + "output": 3072 }, "cost": { - "input": 0.07, - "output": 0.27 + "input": 0.13, + "output": 0 } }, - "qwen/qwen3-235b-a22b-fp8": { - "id": "qwen/qwen3-235b-a22b-fp8", - "name": "Qwen3 235B A22B", + "o1-mini": { + "id": "o1-mini", + "name": "o1-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": false, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-09-12", + "last_updated": "2024-09-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 20000 + "context": 128000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "qwen/qwen3-8b-fp8": { - "id": "qwen/qwen3-8b-fp8", - "name": "Qwen3 8B", + "phi-3.5-mini-instruct": { + "id": "phi-3.5-mini-instruct", + "name": "Phi-3.5-mini-instruct", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", "modalities": { "input": ["text"], "output": ["text"] @@ -39643,2297 +81925,2599 @@ "open_weights": true, "limit": { "context": 128000, - "output": 20000 + "output": 4096 }, "cost": { - "input": 0.035, - "output": 0.138 + "input": 0.13, + "output": 0.52 } }, - "qwen/qwen3-vl-235b-a22b-instruct": { - "id": "qwen/qwen3-vl-235b-a22b-instruct", - "name": "Qwen3 VL 235B A22B Instruct", - "attachment": true, + "mistral-nemo": { + "id": "mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "knowledge": "2024-07", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.15, + "output": 0.15 } }, - "qwen/qwen-2.5-72b-instruct": { - "id": "qwen/qwen-2.5-72b-instruct", - "name": "Qwen 2.5 72B Instruct", - "family": "qwen", - "attachment": false, + "cohere-embed-v-4-0": { + "id": "cohere-embed-v-4-0", + "name": "Embed v4", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-10-15", - "last_updated": "2024-10-15", + "tool_call": false, + "temperature": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 8192 + "context": 128000, + "output": 1536 }, "cost": { - "input": 0.38, - "output": 0.4 + "input": 0.12, + "output": 0 } }, - "qwen/qwen3.5-35b-a3b": { - "id": "qwen/qwen3.5-35b-a3b", - "name": "Qwen3.5-35B-A3B", - "family": "qwen", + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.25, - "output": 2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "kwaipilot/kat-coder-pro": { - "id": "kwaipilot/kat-coder-pro", - "name": "Kat Coder Pro", - "attachment": false, - "reasoning": false, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-01-05", - "last_updated": "2026-01-05", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, + "context": 272000, "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "google/gemma-4-31b-it": { - "id": "google/gemma-4-31b-it", - "name": "Gemma 4 31B", - "family": "gemma", + "gpt-5-chat": { + "id": "gpt-5-chat", + "name": "GPT-5 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "knowledge": "2024-10-24", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.4 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "google/gemma-3-12b-it": { - "id": "google/gemma-3-12b-it", - "name": "Gemma 3 12B", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, + "cohere-command-a": { + "id": "cohere-command-a", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, + "knowledge": "2024-06-01", "release_date": "2025-03-13", "last_updated": "2025-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 256000, + "output": 8000 }, "cost": { - "input": 0.05, - "output": 0.1 + "input": 2.5, + "output": 10 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Gemma 3 27B", - "family": "gemma", + "llama-3.2-11b-vision-instruct": { + "id": "llama-3.2-11b-vision-instruct", + "name": "Llama-3.2-11B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-03-25", - "last_updated": "2025-03-25", + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 98304, - "output": 16384 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.119, - "output": 0.2 + "input": 0.37, + "output": 0.37 } }, - "google/gemma-4-26b-a4b-it": { - "id": "google/gemma-4-26b-a4b-it", - "name": "Gemma 4 26B A4B", - "family": "gemma", + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 400000, + "output": 272000 }, "cost": { - "input": 0.13, - "output": 0.4 + "input": 15, + "output": 120 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, - "reasoning": true, + "cohere-command-r-08-2024": { + "id": "cohere-command-r-08-2024", + "name": "Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.15, + "output": 0.6 } }, - "moonshotai/kimi-k2-instruct": { - "id": "moonshotai/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "attachment": false, + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.57, - "output": 2.3 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "gpt-4": { + "id": "gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2023-11", + "release_date": "2023-03-14", + "last_updated": "2023-03-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 60, + "output": 120 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi", - "attachment": false, + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2025-11-07", - "last_updated": "2025-11-07", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "deepseek/deepseek-v4-flash": { - "id": "deepseek/deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "phi-3-medium-128k-instruct": { + "id": "phi-3-medium-128k-instruct", + "name": "Phi-3-medium-instruct (128k)", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.17, + "output": 0.68 } }, - "deepseek/deepseek-v4-pro": { - "id": "deepseek/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "gpt-4-32k": { + "id": "gpt-4-32k", + "name": "GPT-4 32K", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2023-11", + "release_date": "2023-03-14", + "last_updated": "2023-03-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 393216 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.69, - "output": 3.38, - "cache_read": 0.13 + "input": 60, + "output": 120 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 262144 - }, - "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 - } - } - } - }, - "xiaomi-token-plan-cn": { - "id": "xiaomi-token-plan-cn", - "env": ["XIAOMI_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://token-plan-cn.xiaomimimo.com/v1", - "name": "Xiaomi Token Plan (China)", - "doc": "https://platform.xiaomimimo.com/#/docs", - "models": { - "mimo-v2-tts": { - "id": "mimo-v2-tts", - "name": "MiMo-V2-TTS", - "family": "mimo", + "meta-llama-3.1-405b-instruct": { + "id": "meta-llama-3.1-405b-instruct", + "name": "Meta-Llama-3.1-405B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", - "modalities": { - "input": ["text"], - "output": ["audio"] - }, - "open_weights": true, - "limit": { - "context": 8192, - "output": 16384 - }, - "cost": { - "input": 0, - "output": 0 - } - }, - "mimo-v2.5-pro": { - "id": "mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", - "family": "mimo", - "attachment": false, - "reasoning": true, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 5.33, + "output": 16 } }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo-V2-Omni", - "family": "mimo", - "attachment": true, - "reasoning": true, + "cohere-command-r-plus-08-2024": { + "id": "cohere-command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 131072 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 2.5, + "output": 10 } }, - "mimo-v2.5": { - "id": "mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", - "attachment": true, - "reasoning": true, + "phi-4-mini": { + "id": "phi-4-mini", + "name": "Phi-4-mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.075, + "output": 0.3 } }, - "mimo-v2-pro": { - "id": "mimo-v2-pro", - "name": "MiMo-V2-Pro", - "family": "mimo", + "gpt-3.5-turbo-1106": { + "id": "gpt-3.5-turbo-1106", + "name": "GPT-3.5 Turbo 1106", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2021-08", + "release_date": "2023-11-06", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 1, + "output": 2 } }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "MiMo-V2-Flash", - "family": "mimo", - "attachment": false, - "reasoning": true, + "llama-4-scout-17b-16e-instruct": { + "id": "llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.2, + "output": 0.78 } - } - } - }, - "wandb": { - "id": "wandb", - "env": ["WANDB_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.inference.wandb.ai/v1", - "name": "Weights & Biases", - "doc": "https://docs.wandb.ai/guides/integrations/inference/", - "models": { - "Qwen/Qwen3-30B-A3B-Instruct-2507": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507", - "family": "qwen", + }, + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-29", - "last_updated": "2026-03-12", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.71, + "output": 0.71 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", + "grok-4-20-non-reasoning": { + "id": "grok-4-20-non-reasoning", + "name": "Grok 4.20 (Non-Reasoning)", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-28", - "last_updated": "2026-03-12", + "knowledge": "2025-09", + "release_date": "2026-04-08", + "last_updated": "2026-04-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 262000, + "output": 8192 }, + "status": "beta", "cost": { - "input": 0.1, - "output": 0.1 + "input": 2, + "output": 6 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3-235B-A22B-Thinking-2507", - "family": "qwen", - "attachment": false, + "gpt-5.1-chat": { + "id": "gpt-5.1-chat", + "name": "GPT-5.1 Chat", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "image", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen3-Coder-480B-A35B-Instruct", - "family": "qwen", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek-V4-Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2026-03-12", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 384000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 1, - "output": 1.5 + "input": 1.74, + "output": 3.48 } }, - "zai-org/GLM-5-FP8": { - "id": "zai-org/GLM-5-FP8", - "name": "GLM 5", - "family": "glm", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-03-12", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-12-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1, - "output": 3.2 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "meta-llama/Llama-4-Scout-17B-16E-Instruct": { - "id": "meta-llama/Llama-4-Scout-17B-16E-Instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", - "attachment": false, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-31", - "last_updated": "2026-03-12", + "knowledge": "2025-07-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, + "context": 200000, "output": 64000 }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + }, "cost": { - "input": 0.17, - "output": 0.66 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": false, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 1000000, "output": 128000 }, + "status": "beta", + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + }, "cost": { - "input": 0.71, - "output": 0.71 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "meta-llama/Llama-3.1-8B-Instruct": { - "id": "meta-llama/Llama-3.1-8B-Instruct", - "name": "Meta-Llama-3.1-8B-Instruct", - "family": "llama", + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2026-03-12", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.22, - "output": 0.22 + "input": 1.35, + "output": 5.4 } }, - "meta-llama/Llama-3.1-70B-Instruct": { - "id": "meta-llama/Llama-3.1-70B-Instruct", - "name": "Llama 3.1 70B", - "family": "llama", - "attachment": false, + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "Grok 4.1 Fast (Non-Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2026-03-12", + "release_date": "2025-06-27", + "last_updated": "2025-06-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, + "input": 128000, + "output": 8192 + }, + "status": "beta", + "cost": { + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 Nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.8, - "output": 0.8 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "OpenPipe/Qwen3-14B-Instruct": { - "id": "OpenPipe/Qwen3-14B-Instruct", - "name": "OpenPipe Qwen3 14B Instruct", - "family": "qwen", + "phi-3-small-8k-instruct": { + "id": "phi-3-small-8k-instruct", + "name": "Phi-3-small-instruct (8k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-04-29", - "last_updated": "2026-03-12", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 8192, + "output": 2048 }, "cost": { - "input": 0.05, - "output": 0.22 + "input": 0.15, + "output": 0.6 } }, - "microsoft/Phi-4-mini-instruct": { - "id": "microsoft/Phi-4-mini-instruct", - "name": "Phi-4-mini-instruct", - "family": "phi", - "attachment": false, + "gpt-5.3-chat": { + "id": "gpt-5.3-chat", + "name": "GPT-5.3 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 128000 + "output": 16384 }, "cost": { - "input": 0.08, - "output": 0.35 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8": { - "id": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8", - "name": "NVIDIA Nemotron 3 Super 120B", - "family": "nemotron", + "gpt-3.5-turbo-0125": { + "id": "gpt-3.5-turbo-0125", + "name": "GPT-3.5 Turbo 0125", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-12", + "knowledge": "2021-08", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.5, + "output": 1.5 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "cohere-embed-v3-multilingual": { + "id": "cohere-embed-v3-multilingual", + "name": "Embed v3 Multilingual", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-21", - "last_updated": "2026-03-12", + "tool_call": false, + "temperature": false, + "release_date": "2023-11-07", + "last_updated": "2023-11-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 161000, - "output": 161000 + "context": 512, + "output": 1024 }, "cost": { - "input": 0.55, - "output": 1.65 + "input": 0.1, + "output": 0 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "gpt-oss-20b", - "family": "gpt-oss", + "gpt-3.5-turbo-0613": { + "id": "gpt-3.5-turbo-0613", + "name": "GPT-3.5 Turbo 0613", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2026-03-12", + "knowledge": "2021-08", + "release_date": "2023-06-13", + "last_updated": "2023-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.2 + "input": 3, + "output": 4 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "gpt-oss-120b", - "family": "gpt-oss", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt-codex", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "image", "audio"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 2.85 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax M2.5", - "family": "minimax", + "phi-4-reasoning": { + "id": "phi-4-reasoning", + "name": "Phi-4-reasoning", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-03-12", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 196608 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.125, + "output": 0.5 } }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2025-12-31", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 131072 + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26, - "cache_write": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } - } - } - }, - "chutes": { - "id": "chutes", - "env": ["CHUTES_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://llm.chutes.ai/v1", - "name": "Chutes", - "doc": "https://llm.chutes.ai/v1/models", - "models": { - "NousResearch/DeepHermes-3-Mistral-24B-Preview": { - "id": "NousResearch/DeepHermes-3-Mistral-24B-Preview", - "name": "DeepHermes 3 Mistral 24B Preview", - "family": "nousresearch", - "attachment": false, + }, + "mistral-medium-2505": { + "id": "mistral-medium-2505", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2025-05", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.0245, - "output": 0.0978, - "cache_read": 0.01225 + "input": 0.4, + "output": 2 } }, - "NousResearch/Hermes-4-14B": { - "id": "NousResearch/Hermes-4-14B", - "name": "Hermes 4 14B", - "family": "nousresearch", + "meta-llama-3.1-70b-instruct": { + "id": "meta-llama-3.1-70b-instruct", + "name": "Meta-Llama-3.1-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 40960 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.0136, - "output": 0.0543, - "cache_read": 0.0068 + "input": 2.68, + "output": 3.54 } }, - "Qwen/Qwen3-30B-A3B": { - "id": "Qwen/Qwen3-30B-A3B", - "name": "Qwen3 30B A3B", - "family": "qwen", + "text-embedding-ada-002": { + "id": "text-embedding-ada-002", + "name": "text-embedding-ada-002", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "reasoning": false, + "tool_call": false, + "release_date": "2022-12-15", + "last_updated": "2022-12-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 8192, + "output": 1536 }, "cost": { - "input": 0.06, - "output": 0.22, - "cache_read": 0.03 + "input": 0.1, + "output": 0 } }, - "Qwen/Qwen3-32B-TEE": { - "id": "Qwen/Qwen3-32B-TEE", - "name": "Qwen3 32B TEE", - "family": "qwen", + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.08, - "output": 0.24, - "cache_read": 0.04 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "Qwen/Qwen3.6-27B-TEE": { - "id": "Qwen/Qwen3.6-27B-TEE", - "name": "Qwen3.6 27B TEE", - "family": "qwen", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.195, - "output": 1.56, - "cache_read": 0.0975 + "input": 1.75, + "output": 14, + "cache_read": 0.125 } }, - "Qwen/Qwen2.5-Coder-32B-Instruct": { - "id": "Qwen/Qwen2.5-Coder-32B-Instruct", - "name": "Qwen2.5 Coder 32B Instruct", - "family": "qwen", - "attachment": false, + "gpt-4-turbo-vision": { + "id": "gpt-4-turbo-vision", + "name": "GPT-4 Turbo Vision", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": true, + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.0272, - "output": 0.1087, - "cache_read": 0.0136 + "input": 10, + "output": 30 } }, - "Qwen/Qwen3Guard-Gen-0.6B": { - "id": "Qwen/Qwen3Guard-Gen-0.6B", - "name": "Qwen3Guard Gen 0.6B", - "family": "qwen", + "mistral-large-2411": { + "id": "mistral-large-2411", + "name": "Mistral Large 24.11", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2024-09", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.01, - "output": 0.0109, - "cache_read": 0.005 + "input": 2, + "output": 6 } }, - "Qwen/Qwen3-Next-80B-A3B-Instruct": { - "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "family": "qwen", + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.8, - "cache_read": 0.05 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507-TEE": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507-TEE", - "name": "Qwen3 235B A22B Instruct 2507 TEE", - "family": "qwen", + "text-embedding-3-small": { + "id": "text-embedding-3-small", + "name": "text-embedding-3-small", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "tool_call": false, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 8191, + "output": 1536 }, "cost": { - "input": 0.1, - "output": 0.6, - "cache_read": 0.05 + "input": 0.02, + "output": 0 } }, - "Qwen/Qwen3-Coder-Next-TEE": { - "id": "Qwen/Qwen3-Coder-Next-TEE", - "name": "Qwen3 Coder Next TEE", - "family": "qwen", + "deepseek-v3.2-speciale": { + "id": "deepseek-v3.2-speciale", + "name": "DeepSeek-V3.2-Speciale", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0.75, - "cache_read": 0.06 + "input": 0.58, + "output": 1.68 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3 235B A22B Thinking 2507", - "family": "qwen", + "gpt-3.5-turbo-0301": { + "id": "gpt-3.5-turbo-0301", + "name": "GPT-3.5 Turbo 0301", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2021-08", + "release_date": "2023-03-01", + "last_updated": "2023-03-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0.11, - "output": 0.6, - "cache_read": 0.055 + "input": 1.5, + "output": 2 } }, - "Qwen/Qwen2.5-VL-32B-Instruct": { - "id": "Qwen/Qwen2.5-VL-32B-Instruct", - "name": "Qwen2.5 VL 32B Instruct", - "family": "qwen", - "attachment": true, + "meta-llama-3.1-8b-instruct": { + "id": "meta-llama-3.1-8b-instruct", + "name": "Meta-Llama-3.1-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": true, + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.0543, - "output": 0.2174, - "cache_read": 0.02715 + "input": 0.3, + "output": 0.61 } }, - "Qwen/Qwen3.5-397B-A17B-TEE": { - "id": "Qwen/Qwen3.5-397B-A17B-TEE", - "name": "Qwen3.5 397B A17B TEE", - "family": "qwen", + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-02-18", - "last_updated": "2026-04-25", + "knowledge": "2025-03-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 200000, + "output": 32000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.39, - "output": 2.34, - "cache_read": 0.195 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "Qwen/Qwen2.5-72B-Instruct": { - "id": "Qwen/Qwen2.5-72B-Instruct", - "name": "Qwen2.5 72B Instruct", - "family": "qwen", + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.2989, - "output": 1.1957, - "cache_read": 0.14945 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "zai-org/GLM-5.1-TEE": { - "id": "zai-org/GLM-5.1-TEE", - "name": "GLM 5.1 TEE", - "family": "glm", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "interleaved": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-08", - "last_updated": "2026-04-25", + "knowledge": "2025-01", + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 65535 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 1.05, - "output": 3.5, - "cache_read": 0.525 + "input": 0.6, + "output": 3 } }, - "zai-org/GLM-4.7-FP8": { - "id": "zai-org/GLM-4.7-FP8", - "name": "GLM 4.7 FP8", - "family": "glm", + "phi-4": { + "id": "phi-4", + "name": "Phi-4", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-04-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 65535 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.2989, - "output": 1.1957, - "cache_read": 0.14945 + "input": 0.125, + "output": 0.5 } }, - "zai-org/GLM-5-TEE": { - "id": "zai-org/GLM-5-TEE", - "name": "GLM 5 TEE", - "family": "glm", - "attachment": false, + "grok-4-fast-reasoning": { + "id": "grok-4-fast-reasoning", + "name": "Grok 4 Fast (Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-04-25", + "knowledge": "2025-07", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 65535 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.95, - "output": 2.55, - "cache_read": 0.475 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "zai-org/GLM-4.7-TEE": { - "id": "zai-org/GLM-4.7-TEE", - "name": "GLM 4.7 TEE", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "gpt-image-1.5": { + "id": "gpt-image-1.5", + "name": "GPT-Image-1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-11-25", + "last_updated": "2025-11-25", + "modalities": { + "input": ["text", "image"], + "output": ["text", "image"] }, - "structured_output": true, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + }, + "cost": { + "input": 5, + "output": 32, + "cache_read": 1.25 + } + }, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 65535 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.39, - "output": 1.75, - "cache_read": 0.195 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "zai-org/GLM-4.6V": { - "id": "zai-org/GLM-4.6V", - "name": "GLM 4.6V", - "family": "glm", + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.3, - "output": 0.9, - "cache_read": 0.15 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "zai-org/GLM-5-Turbo": { - "id": "zai-org/GLM-5-Turbo", - "name": "GLM 5 Turbo", - "family": "glm", + "o1": { + "id": "o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 65535 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.4891, - "output": 1.9565, - "cache_read": 0.24455 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "XiaomiMiMo/MiMo-V2-Flash-TEE": { - "id": "XiaomiMiMo/MiMo-V2-Flash-TEE", - "name": "MiMo V2 Flash TEE", - "family": "mimo", + "deepseek-v3.1": { + "id": "deepseek-v3.1", + "name": "DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "knowledge": "2024-07", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.09, - "output": 0.29, - "cache_read": 0.045 + "input": 0.56, + "output": 1.68 } }, - "deepseek-ai/DeepSeek-R1-Distill-Llama-70B": { - "id": "deepseek-ai/DeepSeek-R1-Distill-Llama-70B", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": true, + "llama-3.2-90b-vision-instruct": { + "id": "llama-3.2-90b-vision-instruct", + "name": "Llama-3.2-90B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.0272, - "output": 0.1087, - "cache_read": 0.0136 + "input": 2.04, + "output": 2.04 } }, - "deepseek-ai/DeepSeek-V3.1-TEE": { - "id": "deepseek-ai/DeepSeek-V3.1-TEE", - "name": "DeepSeek V3.1 TEE", - "family": "deepseek", - "attachment": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2025-02-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.27, - "output": 1, - "cache_read": 0.135 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "deepseek-ai/DeepSeek-V3-0324-TEE": { - "id": "deepseek-ai/DeepSeek-V3-0324-TEE", - "name": "DeepSeek V3 0324 TEE", - "family": "deepseek", + "ministral-3b": { + "id": "ministral-3b", + "name": "Ministral 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2024-03", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 1, - "cache_read": 0.125 + "input": 0.04, + "output": 0.04 } }, - "deepseek-ai/DeepSeek-R1-0528-TEE": { - "id": "deepseek-ai/DeepSeek-R1-0528-TEE", - "name": "DeepSeek R1 0528 TEE", - "family": "deepseek-thinking", - "attachment": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 2.15, - "cache_read": 0.225 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "deepseek-ai/DeepSeek-V3.2-TEE": { - "id": "deepseek-ai/DeepSeek-V3.2-TEE", - "name": "DeepSeek V3.2 TEE", - "family": "deepseek", - "attachment": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 Mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.28, - "output": 0.42, - "cache_read": 0.14 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "openai/gpt-oss-120b-TEE": { - "id": "openai/gpt-oss-120b-TEE", - "name": "gpt oss 120b TEE", - "family": "gpt-oss", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "interleaved": true, "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2025-01", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 0.09, - "output": 0.36, - "cache_read": 0.045 + "input": 0.95, + "output": 4 } }, - "unsloth/gemma-3-12b-it": { - "id": "unsloth/gemma-3-12b-it", - "name": "gemma 3 12b it", - "family": "unsloth", + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.03, - "output": 0.1, - "cache_read": 0.015 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "unsloth/Llama-3.2-3B-Instruct": { - "id": "unsloth/Llama-3.2-3B-Instruct", - "name": "Llama 3.2 3B Instruct", - "family": "unsloth", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-02-12", - "last_updated": "2026-04-25", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 200000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.01, - "output": 0.0136, - "cache_read": 0.005 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "unsloth/gemma-3-4b-it": { - "id": "unsloth/gemma-3-4b-it", - "name": "gemma 3 4b it", - "family": "unsloth", + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "Grok 4.1 Fast (Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "release_date": "2025-06-27", + "last_updated": "2025-06-27", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 96000, - "output": 96000 + "context": 128000, + "input": 128000, + "output": 8192 }, + "status": "beta", "cost": { - "input": 0.01, - "output": 0.0272, - "cache_read": 0.005 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "unsloth/Llama-3.2-1B-Instruct": { - "id": "unsloth/Llama-3.2-1B-Instruct", - "name": "Llama 3.2 1B Instruct", - "family": "unsloth", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-04-25", + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 8192 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.01, - "output": 0.0109, - "cache_read": 0.005 + "input": 0.25, + "output": 2, + "cache_read": 0.03 } }, - "unsloth/Mistral-Nemo-Instruct-2407": { - "id": "unsloth/Mistral-Nemo-Instruct-2407", - "name": "Mistral Nemo Instruct 2407", - "family": "unsloth", - "attachment": false, + "gpt-image-1": { + "id": "gpt-image-1", + "name": "GPT-Image-1", + "description": "OpenAI image model for production generation, edits, and brand-safe visual workflows", + "family": "gpt-image", + "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "release_date": "2025-04-24", + "last_updated": "2025-04-24", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 0, + "output": 0 }, "cost": { - "input": 0.02, - "output": 0.04, - "cache_read": 0.01 + "input": 5, + "output": 40, + "cache_read": 1.25 } }, - "unsloth/gemma-3-27b-it": { - "id": "unsloth/gemma-3-27b-it", - "name": "gemma 3 27b it", - "family": "unsloth", + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.0272, - "output": 0.1087, - "cache_read": 0.0136 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "google/gemma-4-31B-turbo-TEE": { - "id": "google/gemma-4-31B-turbo-TEE", - "name": "gemma 4 31B turbo TEE", - "family": "gemma", + "gpt-4-turbo": { + "id": "gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.13, - "output": 0.38, - "cache_read": 0.065 + "input": 10, + "output": 30 } }, - "moonshotai/Kimi-K2.6-TEE": { - "id": "moonshotai/Kimi-K2.6-TEE", - "name": "Kimi K2.6 TEE", - "family": "kimi", + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-04-20", - "last_updated": "2026-04-25", + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65535 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.475 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01 } }, - "moonshotai/Kimi-K2.5-TEE": { - "id": "moonshotai/Kimi-K2.5-TEE", - "name": "Kimi K2.5 TEE", - "family": "kimi", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "phi-3-medium-4k-instruct": { + "id": "phi-3-medium-4k-instruct", + "name": "Phi-3-medium-instruct (4k)", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-01-27", - "last_updated": "2026-04-25", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65535 + "context": 4096, + "output": 1024 }, "cost": { - "input": 0.44, - "output": 2, - "cache_read": 0.22 + "input": 0.17, + "output": 0.68 } }, - "MiniMaxAI/MiniMax-M2.5-TEE": { - "id": "MiniMaxAI/MiniMax-M2.5-TEE", - "name": "MiniMax M2.5 TEE", - "family": "minimax", - "attachment": false, + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-04-25", + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 1.2, - "cache_read": 0.075 + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } } }, - "rednote-hilab/dots.ocr": { - "id": "rednote-hilab/dots.ocr", - "name": "dots.ocr", - "family": "rednote", - "attachment": true, + "cohere-embed-v3-english": { + "id": "cohere-embed-v3-english", + "name": "Embed v3 English", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-04-25", + "temperature": false, + "release_date": "2023-11-07", + "last_updated": "2023-11-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 512, + "output": 1024 }, "cost": { - "input": 0.01, - "output": 0.0109, - "cache_read": 0.005 + "input": 0.1, + "output": 0 } }, - "tngtech/DeepSeek-TNG-R1T2-Chimera-TEE": { - "id": "tngtech/DeepSeek-TNG-R1T2-Chimera-TEE", - "name": "DeepSeek TNG R1T2 Chimera TEE", - "family": "deepseek", - "attachment": false, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-25", - "last_updated": "2026-04-25", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.3, - "output": 1.1, - "cache_read": 0.15 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } - } - } - }, - "dinference": { - "id": "dinference", - "env": ["DINFERENCE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.dinference.com/v1", - "name": "DInference", - "doc": "https://dinference.com", - "models": { - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT OSS 120B", + }, + "codestral-2501": { + "id": "codestral-2501", + "name": "Codestral 25.01", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "codestral", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-08", - "last_updated": "2025-08", + "knowledge": "2024-03", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.0675, - "output": 0.27 + "input": 0.3, + "output": 0.9 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "phi-4-reasoning-plus": { + "id": "phi-4-reasoning-plus", + "name": "Phi-4-reasoning-plus", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.45, - "output": 1.65 + "input": 0.125, + "output": 0.5 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "phi-4-multimodal": { + "id": "phi-4-multimodal", + "name": "Phi-4-multimodal", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "phi", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.75, - "output": 2.4 + "input": 0.08, + "output": 0.32, + "input_audio": 4 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, - "reasoning": true, + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 + } + }, + "model-router": { + "id": "model-router", + "name": "Model Router", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "model-router", + "attachment": true, + "reasoning": false, + "tool_call": true, + "release_date": "2025-05-19", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 3.89 + "input": 0.14, + "output": 0 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "grok-4-20-reasoning": { + "id": "grok-4-20-reasoning", + "name": "Grok 4.20 (Reasoning)", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-09", + "release_date": "2026-04-08", + "last_updated": "2026-04-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 262000, + "output": 8192 }, + "status": "beta", "cost": { - "input": 0.22, - "output": 0.88 + "input": 2, + "output": 6 } - } - } - }, - "vivgrid": { - "id": "vivgrid", - "env": ["VIVGRID_API_KEY"], - "npm": "@ai-sdk/openai", - "api": "https://api.vivgrid.com/v1", - "name": "Vivgrid", - "doc": "https://docs.vivgrid.com/models", - "models": { - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", + }, + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "family": "gpt-codex", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -41946,127 +84530,81 @@ "cost": { "input": 1.25, "output": 10, - "cache_read": 0.125 + "cache_read": 0.13 } }, - "gemini-3.1-flash-lite-preview": { - "id": "gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", - "attachment": true, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "cache_write": 1 + "input": 0.58, + "output": 1.68 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "llama-4-maverick-17b-128e-instruct-fp8": { + "id": "llama-4-maverick-17b-128e-instruct-fp8", + "name": "Llama 4 Maverick 17B 128E Instruct FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", - "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1048576, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" - }, - "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } - } - }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 128000, + "output": 8192 }, "cost": { "input": 0.25, - "output": 2, - "cache_read": 0.03 + "output": 1 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "family": "gpt-codex", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -42082,172 +84620,163 @@ "cache_read": 0.175 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, + "phi-4-mini-reasoning": { + "id": "phi-4-mini-reasoning", + "name": "Phi-4-mini-reasoning", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0.075, + "output": 0.3 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "phi-3-mini-128k-instruct": { + "id": "phi-3-mini-128k-instruct", + "name": "Phi-3-mini-instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.13, + "output": 0.52 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": false, - "reasoning": true, - "tool_call": true, + "gpt-image-2": { + "id": "gpt-image-2", + "name": "GPT-Image-2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 0, + "output": 0 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 5, + "output": 30, + "cache_read": 1.25 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "phi-3-mini-4k-instruct": { + "id": "phi-3-mini-4k-instruct", + "name": "Phi-3-mini-instruct (4k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 4096, + "output": 1024 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.13, + "output": 0.52 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek-V3.2", - "family": "deepseek", + "meta-llama-3-70b-instruct": { + "id": "meta-llama-3-70b-instruct", + "name": "Meta-Llama-3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" + "context": 8192, + "output": 2048 }, "cost": { - "input": 0.28, - "output": 0.42 + "input": 2.68, + "output": 3.54 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", - "attachment": false, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "image", "audio"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 272000, "output": 128000 }, "cost": { @@ -42259,15 +84788,22 @@ "gpt-5.5": { "id": "gpt-5.5", "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -42278,18 +84814,10 @@ "input": 922000, "output": 128000 }, - "provider": { - "npm": "@ai-sdk/openai-compatible" - }, "cost": { "input": 5, "output": 30, "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, "tiers": [ { "input": 10, @@ -42300,214 +84828,105 @@ "size": 272000 } } - ] - } - }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 1000000, - "output": 384000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible" - }, - "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } } } }, - "deepinfra": { - "id": "deepinfra", - "env": ["DEEPINFRA_API_KEY"], - "npm": "@ai-sdk/deepinfra", - "name": "Deep Infra", - "doc": "https://deepinfra.com/models", + "cerebras": { + "id": "cerebras", + "env": ["CEREBRAS_API_KEY"], + "npm": "@ai-sdk/cerebras", + "name": "Cerebras", + "doc": "https://inference-docs.cerebras.ai/models/overview", "models": { - "Qwen/Qwen3.5-397B-A17B": { - "id": "Qwen/Qwen3.5-397B-A17B", - "name": "Qwen 3.5 397B A17B", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-01", - "last_updated": "2026-04-20", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 81920 - }, - "cost": { - "input": 0.54, - "output": 3.4 - } - }, - "Qwen/Qwen3.5-35B-A3B": { - "id": "Qwen/Qwen3.5-35B-A3B", - "name": "Qwen 3.5 35B A3B", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-01", - "last_updated": "2026-04-20", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 81920 - }, - "cost": { - "input": 0.2, - "output": 0.95 - } - }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo", - "name": "Qwen3 Coder 480B A35B Instruct Turbo", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 66536 - }, - "cost": { - "input": 0.3, - "output": 1.2 - } - }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 66536 - }, - "cost": { - "input": 0.4, - "output": 1.6 - } - }, - "Qwen/Qwen3.6-35B-A3B": { - "id": "Qwen/Qwen3.6-35B-A3B", - "name": "Qwen3.6 35B A3B", - "family": "qwen", + "gemma-4-31b": { + "id": "gemma-4-31b", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "release_date": "2026-04-02", + "last_updated": "2026-07-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 81920 + "context": 131072, + "output": 40960 }, + "status": "beta", "cost": { - "input": 0.2, - "output": 1 + "input": 0.99, + "output": 1.49 } }, - "zai-org/GLM-4.7-Flash": { - "id": "zai-org/GLM-4.7-Flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-08-05", + "last_updated": "2026-06-10", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 16384 + "context": 131072, + "output": 40960 }, "cost": { - "input": 0.06, - "output": 0.4 + "input": 0.35, + "output": 0.75 } }, - "zai-org/GLM-4.5": { - "id": "zai-org/GLM-4.5", - "name": "GLM-4.5", - "family": "glm", + "zai-glm-4.7": { + "id": "zai-glm-4.7", + "name": "Z.AI GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-01-07", + "last_updated": "2026-06-10", "modalities": { "input": ["text"], "output": ["text"] @@ -42515,20 +84934,38 @@ "open_weights": true, "limit": { "context": 131072, - "output": 98304 + "output": 40960 }, - "status": "deprecated", + "status": "beta", "cost": { - "input": 0.6, - "output": 2.2 + "input": 2.25, + "output": 2.75, + "cache_read": 0, + "cache_write": 0 } - }, - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", + } + } + }, + "zai-coding-plan": { + "id": "zai-coding-plan", + "env": ["ZHIPU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.z.ai/api/coding/paas/v4", + "name": "Z.AI Coding Plan", + "doc": "https://docs.z.ai/devpack/overview", + "models": { + "glm-4.7": { + "id": "glm-4.7", "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -42543,465 +84980,552 @@ }, "open_weights": true, "limit": { - "context": 202752, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.43, - "output": 1.75, - "cache_read": 0.08 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", + "glm-5.1": { + "id": "glm-5.1", "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 200000, + "output": 131072 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 16384 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.8, - "output": 2.56, - "cache_read": 0.16 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/GLM-4.6V": { - "id": "zai-org/GLM-4.6V", - "name": "GLM-4.6V", + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", "family": "glm", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, + "context": 200000, "output": 131072 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "GLM-4.6", - "family": "glm", + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.43, - "output": 1.74, - "cache_read": 0.08 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "meta-llama/Llama-4-Scout-17B-16E-Instruct": { - "id": "meta-llama/Llama-4-Scout-17B-16E-Instruct", - "name": "Llama 4 Scout 17B", - "family": "llama", + "glm-5-turbo": { + "id": "glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 10000000, - "output": 16384 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.08, - "output": 0.3 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "meta-llama/Llama-3.1-8B-Instruct": { - "id": "meta-llama/Llama-3.1-8B-Instruct", - "name": "Llama 3.1 8B", - "family": "llama", + } + } + }, + "nvidia": { + "id": "nvidia", + "env": ["NVIDIA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://integrate.api.nvidia.com/v1", + "name": "Nvidia", + "doc": "https://docs.api.nvidia.com/nim/", + "models": { + "baai/bge-m3": { + "id": "baai/bge-m3", + "name": "BGE M3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "bge", "attachment": false, "reasoning": false, - "tool_call": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": false, + "temperature": false, + "release_date": "2024-01-30", + "last_updated": "2026-04-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 8192, + "output": 1024 }, "cost": { - "input": 0.02, - "output": 0.05 + "input": 0, + "output": 0 } }, - "meta-llama/Llama-3.1-70B-Instruct": { - "id": "meta-llama/Llama-3.1-70B-Instruct", - "name": "Llama 3.1 70B", - "family": "llama", + "moonshotai/kimi-k2-instruct-0905": { + "id": "moonshotai/kimi-k2-instruct-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.4, - "output": 0.4 + "input": 0, + "output": 0 } }, - "meta-llama/Llama-3.1-8B-Instruct-Turbo": { - "id": "meta-llama/Llama-3.1-8B-Instruct-Turbo", - "name": "Llama 3.1 8B Turbo", - "family": "llama", - "attachment": false, - "reasoning": false, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.02, - "output": 0.03 + "input": 0, + "output": 0 } }, - "meta-llama/Llama-3.3-70B-Instruct-Turbo": { - "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo", - "name": "Llama 3.3 70B Turbo", - "family": "llama", - "attachment": false, - "reasoning": false, + "minimaxai/minimax-m3": { + "id": "minimaxai/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 1000000, "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.32 + "input": 0, + "output": 0 } }, - "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { - "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", - "name": "Llama 4 Maverick 17B FP8", - "family": "llama", + "minimaxai/minimax-m2.7": { + "id": "minimaxai/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0, + "output": 0 } }, - "meta-llama/Llama-3.1-70B-Instruct-Turbo": { - "id": "meta-llama/Llama-3.1-70B-Instruct-Turbo", - "name": "Llama 3.1 70B Turbo", - "family": "llama", - "attachment": false, - "reasoning": false, + "stepfun-ai/step-3.7-flash": { + "id": "stepfun-ai/step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "temperature": true, + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 256000, "output": 16384 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek-R1-0528", + "stepfun-ai/step-3.5-flash": { + "id": "stepfun-ai/step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, "reasoning": true, - "tool_call": false, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "release_date": "2026-02-02", + "last_updated": "2026-02-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 64000 + "context": 256000, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 2.15, - "cache_read": 0.35 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek-V3.2", - "attachment": false, - "reasoning": true, + "google/gemma-3n-e4b-it": { + "id": "google/gemma-3n-e4b-it", + "name": "Gemma 3n E4b It", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "knowledge": "2024-06", + "release_date": "2025-06-03", + "last_updated": "2025-06-03", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 64000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.26, - "output": 0.38, - "cache_read": 0.13 + "input": 0, + "output": 0 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, + "google/gemma-3n-e2b-it": { + "id": "google/gemma-3n-e2b-it", + "name": "Gemma 3n E2b It", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-06", + "release_date": "2025-06-12", + "last_updated": "2025-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.03, - "output": 0.14 + "input": 0, + "output": 0 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, - "tool_call": true, + "google/google-paligemma": { + "id": "google/google-paligemma", + "name": "paligemma", + "description": "Gemini multimodal model for text, image, audio, video, and document tasks", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-05-14", + "last_updated": "2024-08-26", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.24 + "input": 0, + "output": 0 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma-4-31B-IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-11-06", - "last_updated": "2025-11-07", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 256000, + "output": 16384 }, "cost": { - "input": 0.47, - "output": 2 + "input": 0, + "output": 0 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "Kimi K2.6", - "family": "kimi", - "attachment": true, - "reasoning": true, + "google/gemma-2-2b-it": { + "id": "google/gemma-2-2b-it", + "name": "Gemma 2 2b It", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2024-07-16", + "last_updated": "2024-07-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.75, - "output": 3.5, - "cache_read": 0.15 + "input": 0, + "output": 0 } }, - "moonshotai/Kimi-K2-Instruct": { - "id": "moonshotai/Kimi-K2-Instruct", - "name": "Kimi K2", - "family": "kimi", + "microsoft/phi-4-mini-instruct": { + "id": "microsoft/phi-4-mini-instruct", + "name": "Phi-4-Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "knowledge": "2024-12", + "release_date": "2024-12-01", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] @@ -43009,624 +85533,644 @@ "open_weights": true, "limit": { "context": 131072, - "output": 32768 + "output": 8192 }, "cost": { - "input": 0.5, - "output": 2 + "input": 0, + "output": 0 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "microsoft/phi-4-multimodal-instruct": { + "id": "microsoft/phi-4-multimodal-instruct", + "name": "Phi 4 Multimodal", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.15 + "input": 0, + "output": 0 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, + "z-ai/glm-5.2": { + "id": "z-ai/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.5, - "output": 2.8 + "input": 0, + "output": 0 } }, - "MiniMaxAI/MiniMax-M2": { - "id": "MiniMaxAI/MiniMax-M2", - "name": "MiniMax M2", - "family": "minimax", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT-OSS-120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-08", + "release_date": "2025-08-04", + "last_updated": "2025-08-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.254, - "output": 1.02 + "input": 0, + "output": 0 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax M2.5", - "family": "minimax", + "openai/whisper-large-v3": { + "id": "openai/whisper-large-v3", + "name": "Whisper Large v3", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2023-09-01", + "last_updated": "2025-09-05", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 0, + "output": 4096 }, "cost": { - "input": 0.27, - "output": 0.95, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 0, + "output": 0 } }, - "MiniMaxAI/MiniMax-M2.1": { - "id": "MiniMaxAI/MiniMax-M2.1", - "name": "MiniMax M2.1", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 196608 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.28, - "output": 1.2 + "input": 0, + "output": 0 } }, - "anthropic/claude-3-7-sonnet-latest": { - "id": "anthropic/claude-3-7-sonnet-latest", - "name": "Claude Sonnet 3.7 (Latest)", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "bytedance/seed-oss-36b-instruct": { + "id": "bytedance/seed-oss-36b-instruct", + "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "seed", + "attachment": false, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "release_date": "2025-09-04", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 3.3, - "output": 16.5, - "cache_read": 0.33 + "input": 0, + "output": 0 } }, - "anthropic/claude-4-opus": { - "id": "anthropic/claude-4-opus", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "mistralai/mistral-7b-instruct-v03": { + "id": "mistralai/mistral-7b-instruct-v03", + "name": "Mistral-7B-Instruct-v0.3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": false, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-06-12", - "last_updated": "2025-06-12", + "release_date": "2025-04-01", + "last_updated": "2025-04-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 16.5, - "output": 82.5 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-V4-Flash": { - "id": "deepseek-ai/DeepSeek-V4-Flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "mistralai/magistral-small-2506": { + "id": "mistralai/magistral-small-2506", + "name": "Magistral Small 2506", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0, + "output": 0 } }, - "deepseek-ai/DeepSeek-V4-Pro": { - "id": "deepseek-ai/DeepSeek-V4-Pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "mistralai/mixtral-8x7b-instruct": { + "id": "mistralai/mixtral-8x7b-instruct", + "name": "Mistral: Mixtral 8x7B Instruct", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2023-12-10", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 32768, + "output": 16384 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0, + "output": 0 } }, - "google/gemma-4-26B-A4B-it": { - "id": "google/gemma-4-26B-A4B-it", - "name": "Gemma 4 26B", - "family": "gemma", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "mistralai/mistral-medium-3-instruct": { + "id": "mistralai/mistral-medium-3-instruct", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 131072, + "input": 131072, + "output": 32768 }, "cost": { - "input": 0.07, - "output": 0.34 + "input": 0, + "output": 0 } }, - "google/gemma-4-31B-it": { - "id": "google/gemma-4-31B-it", - "name": "Gemma 4 31B", - "family": "gemma", - "attachment": false, + "mistralai/mistral-small-4-119b-2603": { + "id": "mistralai/mistral-small-4-119b-2603", + "name": "mistral-small-4-119b-2603", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, + "context": 128000, "output": 8192 }, "cost": { - "input": 0.13, - "output": 0.38 + "input": 0, + "output": 0 } }, - "xiaomi/mimo-v2.5-pro": { - "id": "xiaomi/mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", - "family": "mimo", + "mistralai/mistral-nemotron": { + "id": "mistralai/mistral-nemotron", + "name": "mistral-nemotron", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "nemotron", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2025-06-11", + "last_updated": "2025-06-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 16384 + "context": 128000, + "output": 8192 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0, + "output": 0 } }, - "xiaomi/mimo-v2.5": { - "id": "xiaomi/mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", + "mistralai/mistral-large-3-675b-instruct-2512": { + "id": "mistralai/mistral-large-3-675b-instruct-2512", + "name": "Mistral Large 3 675B Instruct 2512", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "knowledge": "2025-01", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 16384 + "output": 262144 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - }, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0, + "output": 0 } - } - } - }, - "qiniu-ai": { - "id": "qiniu-ai", - "env": ["QINIU_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.qnaigc.com/v1", - "name": "Qiniu", - "doc": "https://developer.qiniu.com/aitokenapi", - "models": { - "qwen3-235b-a22b": { - "id": "qwen3-235b-a22b", - "name": "Qwen 3 235B A22B", + }, + "mistralai/mixtral-8x22b-instruct": { + "id": "mistralai/mixtral-8x22b-instruct", + "name": "Mistral: Mixtral 8x22B Instruct", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-04-17", + "last_updated": "2024-04-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 65536, + "output": 13108 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-1.6-flash": { - "id": "doubao-seed-1.6-flash", - "name": "Doubao-Seed 1.6 Flash", + "nvidia/cosmos-transfer1-7b": { + "id": "nvidia/cosmos-transfer1-7b", + "name": "cosmos-transfer1-7b", + "description": "Video model for prompt-guided generation, editing, and motion workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-15", - "last_updated": "2025-08-15", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-06-13", + "last_updated": "2025-06-30", "modalities": { "input": ["text", "image", "video"], - "output": ["text"] + "output": ["video"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235b A22B Instruct 2507", - "attachment": false, + "nvidia/cosmos-transfer2_5-2b": { + "id": "nvidia/cosmos-transfer2_5-2b", + "name": "cosmos-transfer2.5-2b", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "tool_call": false, + "temperature": false, + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], + "output": ["video"] + }, + "open_weights": true, + "limit": { + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "nvidia/llama-nemotron-embed-vl-1b-v2": { + "id": "nvidia/llama-nemotron-embed-vl-1b-v2", + "name": "llama-nemotron-embed-vl-1b-v2", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "nemotron", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-02-10", + "last_updated": "2026-02-10", + "modalities": { + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 64000 + "context": 32768, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-2.0-code": { - "id": "doubao-seed-2.0-code", - "name": "Doubao Seed 2.0 Code", + "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": { + "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning", + "name": "Nemotron 3 Nano Omni", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": -1, + "max": 32768 + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 256000, - "output": 128000 + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek-v3-0324": { - "id": "deepseek-v3-0324", - "name": "DeepSeek-V3-0324", - "attachment": false, + "nvidia/magpie-tts-zeroshot": { + "id": "nvidia/magpie-tts-zeroshot", + "name": "magpie-tts-zeroshot", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "tool_call": false, + "temperature": false, + "release_date": "2025-05-22", + "last_updated": "2025-06-12", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["audio"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16000 + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-1.5-thinking-pro": { - "id": "doubao-1.5-thinking-pro", - "name": "Doubao 1.5 Thinking Pro", + "nvidia/nvidia-nemotron-nano-9b-v2": { + "id": "nvidia/nvidia-nemotron-nano-9b-v2", + "name": "nvidia-nemotron-nano-9b-v2", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-09", + "release_date": "2025-08-18", + "last_updated": "2025-08-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16000 - } - }, - "claude-3.7-sonnet": { - "id": "claude-3.7-sonnet", - "name": "Claude 3.7 Sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", - "modalities": { - "input": ["text", "image"], - "output": ["text"] + "context": 131072, + "output": 131072 }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 128000 + "cost": { + "input": 0, + "output": 0 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", + "nvidia/synthetic-video-detector": { + "id": "nvidia/synthetic-video-detector", + "name": "synthetic-video-detector", + "description": "Video model for prompt-guided generation, editing, and motion workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-02-22", - "last_updated": "2026-02-22", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen-vl-max-2025-01-25": { - "id": "qwen-vl-max-2025-01-25", - "name": "Qwen VL-MAX-2025-01-25", - "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "nvidia/nemotron-content-safety-reasoning-4b": { + "id": "nvidia/nemotron-content-safety-reasoning-4b", + "name": "nemotron-content-safety-reasoning-4b", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "nemotron", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", + "nvidia/nv-embed-v1": { + "id": "nvidia/nv-embed-v1", + "name": "nv-embed-v1", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-06-07", + "last_updated": "2025-07-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 40000, - "output": 4096 + "context": 32768, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-1.5-pro-32k": { - "id": "doubao-1.5-pro-32k", - "name": "Doubao 1.5 Pro 32k", + "nvidia/usdcode": { + "id": "nvidia/usdcode", + "name": "usdcode", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-01-01", + "last_updated": "2026-01-01", "modalities": { "input": ["text"], "output": ["text"] @@ -43634,2924 +86178,4096 @@ "open_weights": false, "limit": { "context": 128000, - "output": 12000 + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen2.5-vl-72b-instruct": { - "id": "qwen2.5-vl-72b-instruct", - "name": "Qwen 2.5 VL 72B Instruct", - "attachment": true, + "nvidia/riva-translate-4b-instruct-v1_1": { + "id": "nvidia/riva-translate-4b-instruct-v1_1", + "name": "riva-translate-4b-instruct-v1_1", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "tool_call": false, + "temperature": false, + "release_date": "2025-12-12", + "last_updated": "2025-12-12", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 8192 + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-2.0-flash": { - "id": "gemini-2.0-flash", - "name": "Gemini 2.0 Flash", + "nvidia/sparsedrive": { + "id": "nvidia/sparsedrive", + "name": "sparsedrive", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-03-18", + "last_updated": "2025-07-20", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, + "context": 128000, "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-vl-30b-a3b-thinking": { - "id": "qwen3-vl-30b-a3b-thinking", - "name": "Qwen3-Vl 30b A3b Thinking", - "attachment": true, + "nvidia/rerank-qa-mistral-4b": { + "id": "nvidia/rerank-qa-mistral-4b", + "name": "rerank-qa-mistral-4b", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2026-02-09", - "last_updated": "2026-02-09", + "tool_call": false, + "temperature": false, + "release_date": "2024-03-17", + "last_updated": "2025-01-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-3.0-pro-image-preview": { - "id": "gemini-3.0-pro-image-preview", - "name": "Gemini 3.0 Pro Image Preview", + "nvidia/streampetr": { + "id": "nvidia/streampetr", + "name": "streampetr", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, "temperature": true, - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["video"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, + "context": 128000, "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", + "nvidia/active-speaker-detection": { + "id": "nvidia/active-speaker-detection", + "name": "Active Speaker Detection", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.5-opus": { - "id": "claude-4.5-opus", - "name": "Claude 4.5 Opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "nvidia/llama-3_1-nemotron-safety-guard-8b-v3": { + "id": "nvidia/llama-3_1-nemotron-safety-guard-8b-v3", + "name": "llama-3.1-nemotron-safety-guard-8b-v3", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "nemotron", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek-R1", + "nvidia/llama-3_2-nemoretriever-300m-embed-v1": { + "id": "nvidia/llama-3_2-nemoretriever-300m-embed-v1", + "name": "llama-3_2-nemoretriever-300m-embed-v1", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-07-24", + "last_updated": "2025-07-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 32768, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.0-opus": { - "id": "claude-4.0-opus", - "name": "Claude 4.0 Opus", + "nvidia/nemotron-voicechat": { + "id": "nvidia/nemotron-voicechat", + "name": "nemotron-voicechat", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "family": "nemotron", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.5-haiku": { - "id": "claude-4.5-haiku", - "name": "Claude 4.5 Haiku", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-10-16", - "last_updated": "2025-10-16", + "nvidia/nv-embedcode-7b-v1": { + "id": "nvidia/nv-embedcode-7b-v1", + "name": "nv-embedcode-7b-v1", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-03-17", + "last_updated": "2025-05-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 32768, + "output": 2048 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3 Max", + "nvidia/nemotron-3-ultra-550b-a55b": { + "id": "nvidia/nemotron-3-ultra-550b-a55b", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, + "context": 1000000, "output": 65536 + }, + "cost": { + "input": 0.5, + "output": 2.5, + "cache_read": 0.15 } }, - "gemini-3.0-flash-preview": { - "id": "gemini-3.0-flash-preview", - "name": "Gemini 3.0 Flash Preview", - "attachment": true, + "nvidia/nemotron-3-nano-30b-a3b": { + "id": "nvidia/nemotron-3-nano-30b-a3b", + "name": "nemotron-3-nano-30b-a3b", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "knowledge": "2024-09", + "release_date": "2024-12", + "last_updated": "2024-12", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-2.5-flash-image": { - "id": "gemini-2.5-flash-image", - "name": "Gemini 2.5 Flash Image", + "nvidia/cosmos-predict1-5b": { + "id": "nvidia/cosmos-predict1-5b", + "name": "cosmos-predict1-5b", + "description": "Video model for prompt-guided generation, editing, and motion workflows", "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2025-10-22", - "last_updated": "2025-10-22", + "temperature": false, + "release_date": "2025-03-18", + "last_updated": "2025-03-18", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text", "image", "video"], + "output": ["video"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "output": 8192 + "context": 0, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM 4.5", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, + "nvidia/bevformer": { + "id": "nvidia/bevformer", + "name": "bevformer", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-03-18", + "last_updated": "2025-07-20", "modalities": { - "input": ["text"], + "input": ["video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-3.5-sonnet": { - "id": "claude-3.5-sonnet", - "name": "Claude 3.5 Sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, + "nvidia/studiovoice": { + "id": "nvidia/studiovoice", + "name": "studiovoice", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-09-09", - "last_updated": "2025-09-09", + "release_date": "2024-10-03", + "last_updated": "2025-06-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8200 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.0-sonnet": { - "id": "claude-4.0-sonnet", - "name": "Claude 4.0 Sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, + "nvidia/gliner-pii": { + "id": "nvidia/gliner-pii", + "name": "gliner-pii", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-30b-a3b-instruct-2507": { - "id": "qwen3-30b-a3b-instruct-2507", - "name": "Qwen3 30b A3b Instruct 2507", + "nvidia/nemotron-mini-4b-instruct": { + "id": "nvidia/nemotron-mini-4b-instruct", + "name": "nemotron-mini-4b-instruct", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-02-04", - "last_updated": "2026-02-04", + "release_date": "2024-08-21", + "last_updated": "2024-08-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 32000 + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-1.6-thinking": { - "id": "doubao-seed-1.6-thinking", - "name": "Doubao-Seed 1.6 Thinking", + "nvidia/llama-nemotron-rerank-vl-1b-v2": { + "id": "nvidia/llama-nemotron-rerank-vl-1b-v2", + "name": "llama-nemotron-rerank-vl-1b-v2", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "nemotron", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-15", - "last_updated": "2025-08-15", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-03-31", + "last_updated": "2026-03-31", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "attachment": true, + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "Nemotron 3 Super", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-04", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 64000 + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.2, + "output": 0.8 } }, - "qwen3-235b-a22b-thinking-2507": { - "id": "qwen3-235b-a22b-thinking-2507", - "name": "Qwen3 235B A22B Thinking 2507", + "nvidia/usdvalidate": { + "id": "nvidia/usdvalidate", + "name": "usdvalidate", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-07-24", + "last_updated": "2025-01-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, + "context": 0, "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-next-80b-a3b-thinking": { - "id": "qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", + "nvidia/nemotron-3-content-safety": { + "id": "nvidia/nemotron-3-content-safety", + "name": "nemotron-3-content-safety", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "nemotron", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-12", - "last_updated": "2025-09-12", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-30b-a3b-thinking-2507": { - "id": "qwen3-30b-a3b-thinking-2507", - "name": "Qwen3 30b A3b Thinking 2507", + "abacusai/dracarys-llama-3_1-70b-instruct": { + "id": "abacusai/dracarys-llama-3_1-70b-instruct", + "name": "dracarys-llama-3.1-70b-instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-02-04", - "last_updated": "2026-02-04", + "release_date": "2024-09-11", + "last_updated": "2025-05-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 126000, - "output": 32000 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM 4.5 Air", + "deepseek-ai/deepseek-v4-flash": { + "id": "deepseek-ai/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high", "max"] + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 4096 + "context": 1048576, + "output": 393216 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "deepseek-v3.1": { - "id": "deepseek-v3.1", - "name": "DeepSeek-V3.1", + "deepseek-ai/deepseek-v4-pro": { + "id": "deepseek-ai/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high", "max"] + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 1048576, + "output": 393216 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "qwen3-30b-a3b": { - "id": "qwen3-30b-a3b", - "name": "Qwen3 30B A3B", + "qwen/qwen3-next-80b-a3b-instruct": { + "id": "qwen/qwen3-next-80b-a3b-instruct", + "name": "Qwen3-Next-80B-A3B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-12", + "release_date": "2024-12-01", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 40000, - "output": 4096 + "context": 262144, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.1-opus": { - "id": "claude-4.1-opus", - "name": "Claude 4.1 Opus", + "qwen/qwen-image-edit": { + "id": "qwen/qwen-image-edit", + "name": "Qwen Image Edit", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": false, "temperature": true, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "release_date": "2025-08-19", + "last_updated": "2025-08-19", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 0, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-2.0-mini": { - "id": "doubao-seed-2.0-mini", - "name": "Doubao Seed 2.0 Mini", + "qwen/qwen-image": { + "id": "qwen/qwen-image", + "name": "Qwen Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": false, "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 32000 + "context": 0, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-next-80b-a3b-instruct": { - "id": "qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", + "qwen/qwen3-coder-480b-a35b-instruct": { + "id": "qwen/qwen3-coder-480b-a35b-instruct", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-09-12", - "last_updated": "2025-09-12", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - } - }, - "doubao-seed-1.6": { - "id": "doubao-seed-1.6", - "name": "Doubao-Seed 1.6", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-15", - "last_updated": "2025-08-15", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] + "context": 262144, + "output": 66536 }, - "open_weights": false, - "limit": { - "context": 256000, - "output": 32000 + "cost": { + "input": 0, + "output": 0 } }, - "qwen2.5-vl-7b-instruct": { - "id": "qwen2.5-vl-7b-instruct", - "name": "Qwen 2.5 VL 7B Instruct", - "attachment": true, + "qwen/qwen2.5-coder-32b-instruct": { + "id": "qwen/qwen2.5-coder-32b-instruct", + "name": "Qwen2.5 Coder 32b Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-11-06", + "last_updated": "2024-11-06", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 8192 - } - }, - "kling-v2-6": { - "id": "kling-v2-6", - "name": "Kling-V2 6", - "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2026-01-13", - "last_updated": "2026-01-13", - "modalities": { - "input": ["text", "image", "video"], - "output": ["video"] + "output": 4096 }, - "open_weights": false, - "limit": { - "context": 99999999, - "output": 99999999 + "cost": { + "input": 0, + "output": 0 } }, - "MiniMax-M1": { - "id": "MiniMax-M1", - "name": "MiniMax M1", - "attachment": false, + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5-397B-A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2026-01", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 80000 + "context": 262144, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-3.0-pro-preview": { - "id": "gemini-3.0-pro-preview", - "name": "Gemini 3.0 Pro Preview", + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image", "video", "pdf", "audio"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-2.0-lite": { - "id": "doubao-seed-2.0-lite", - "name": "Doubao Seed 2.0 Lite", - "attachment": true, - "reasoning": true, + "sarvamai/sarvam-m": { + "id": "sarvamai/sarvam-m", + "name": "sarvam-m", + "description": "Efficient Indian-language reasoning model for chat, coding, and multilingual work", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-coder-480b-a35b-instruct": { - "id": "qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", + "meta/llama-3.1-8b-instruct": { + "id": "meta/llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-14", - "last_updated": "2025-08-14", + "knowledge": "2023-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, + "context": 16000, "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-3.5-haiku": { - "id": "claude-3.5-haiku", - "name": "Claude 3.5 Haiku", - "attachment": true, + "meta/llama-3.1-70b-instruct": { + "id": "meta/llama-3.1-70b-instruct", + "name": "Llama 3.1 70b Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "release_date": "2024-07-16", + "last_updated": "2024-07-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "gpt-oss-20b", + "meta/llama-3.2-1b-instruct": { + "id": "meta/llama-3.2-1b-instruct", + "name": "Llama 3.2 1b Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "knowledge": "2023-12", + "release_date": "2024-09-18", + "last_updated": "2024-09-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen-turbo": { - "id": "qwen-turbo", - "name": "Qwen-Turbo", - "attachment": false, - "reasoning": true, + "meta/llama-3.2-11b-vision-instruct": { + "id": "meta/llama-3.2-11b-vision-instruct", + "name": "Llama 3.2 11b Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "attachment": true, + "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2023-12", + "release_date": "2024-09-18", + "last_updated": "2024-09-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, + "context": 128000, "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "kimi-k2": { - "id": "kimi-k2", - "name": "Kimi K2", + "meta/llama-3.3-70b-instruct": { + "id": "meta/llama-3.3-70b-instruct", + "name": "Llama 3.3 70b Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-11-26", + "last_updated": "2024-11-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 128000 + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", + "meta/llama-guard-4-12b": { + "id": "meta/llama-guard-4-12b", + "name": "Llama Guard 4 12B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-04-05", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 64000 + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen3-max-preview": { - "id": "qwen3-max-preview", - "name": "Qwen3 Max Preview", + "meta/esmfold": { + "id": "meta/esmfold", + "name": "esmfold", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, + "tool_call": false, "temperature": true, - "release_date": "2025-09-06", - "last_updated": "2025-09-06", + "release_date": "2024-03-15", + "last_updated": "2025-06-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "gpt-oss-120b", + "meta/esm2-650m": { + "id": "meta/esm2-650m", + "name": "esm2-650m", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "release_date": "2024-08-29", + "last_updated": "2025-03-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 4096 + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-1.5-vision-pro": { - "id": "doubao-1.5-vision-pro", - "name": "Doubao 1.5 Vision Pro", + "meta/llama-3.2-90b-vision-instruct": { + "id": "meta/llama-3.2-90b-vision-instruct", + "name": "Llama-3.2-90B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, - "structured_output": false, + "tool_call": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16000 + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "claude-4.5-sonnet": { - "id": "claude-4.5-sonnet", - "name": "Claude 4.5 Sonnet", + "meta/llama-4-maverick-17b-128e-instruct": { + "id": "meta/llama-4-maverick-17b-128e-instruct", + "name": "Llama 4 Maverick 17b 128e Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2024-02", + "release_date": "2025-04-01", + "last_updated": "2025-04-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek-v3": { - "id": "deepseek-v3", - "name": "DeepSeek-V3", + "meta/llama-3.2-3b-instruct": { + "id": "meta/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-08-13", + "release_date": "2024-09-18", + "last_updated": "2024-09-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16000 + "context": 32768, + "output": 32000 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek-R1-0528", + "upstage/solar-10_7b-instruct": { + "id": "upstage/solar-10_7b-instruct", + "name": "solar-10.7b-instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-06-05", + "last_updated": "2025-04-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 32000 - } - }, - "gemini-2.0-flash-lite": { - "id": "gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1048576, "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 } }, - "qwen-max-2025-01-25": { - "id": "qwen-max-2025-01-25", - "name": "Qwen2.5-Max-2025-01-25", + "black-forest-labs/flux_1-schnell": { + "id": "black-forest-labs/flux_1-schnell", + "name": "FLUX.1-schnell", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "temperature": false, + "knowledge": "2024-07", + "release_date": "2024-08-01", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 77, + "input": 77, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 } }, - "doubao-seed-2.0-pro": { - "id": "doubao-seed-2.0-pro", - "name": "Doubao Seed 2.0 Pro", + "black-forest-labs/flux_1-kontext-dev": { + "id": "black-forest-labs/flux_1-kontext-dev", + "name": "FLUX.1-Kontext-dev", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-08-12", + "last_updated": "2025-08-12", "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 40960, + "output": 40960 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek/deepseek-v3.2-exp-thinking": { - "id": "deepseek/deepseek-v3.2-exp-thinking", - "name": "DeepSeek/DeepSeek-V3.2-Exp-Thinking", + "black-forest-labs/flux.1-dev": { + "id": "black-forest-labs/flux.1-dev", + "name": "FLUX.1-dev", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2024-08", + "release_date": "2024-08-01", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32000 + "context": 4096, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 } }, - "deepseek/deepseek-v3.1-terminus-thinking": { - "id": "deepseek/deepseek-v3.1-terminus-thinking", - "name": "DeepSeek/DeepSeek-V3.1-Terminus-Thinking", + "black-forest-labs/flux_2-klein-4b": { + "id": "black-forest-labs/flux_2-klein-4b", + "name": "FLUX.2 Klein 4B", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "flux", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, "temperature": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "knowledge": "2025-06", + "release_date": "2026-01-14", + "last_updated": "2026-01-31", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["image", "text"], + "output": ["image"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 40960, + "output": 40960 + }, + "cost": { + "input": 0, + "output": 0 } - }, - "deepseek/deepseek-v3.2-exp": { - "id": "deepseek/deepseek-v3.2-exp", - "name": "DeepSeek/DeepSeek-V3.2-Exp", - "attachment": false, - "reasoning": false, + } + } + }, + "evroc": { + "id": "evroc", + "env": ["EVROC_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://models.think.evroc.com/v1", + "name": "evroc", + "doc": "https://docs.evroc.com/products/think/overview.html", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 1.4375, + "output": 5.75 } }, - "deepseek/deepseek-v3.2-251201": { - "id": "deepseek/deepseek-v3.2-251201", - "name": "Deepseek/DeepSeek-V3.2", - "attachment": false, + "google/gemma-4-26B-A4B-it": { + "id": "google/gemma-4-26B-A4B-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0.144, + "output": 0.575 } }, - "deepseek/deepseek-math-v2": { - "id": "deepseek/deepseek-math-v2", - "name": "Deepseek/Deepseek-Math-V2", + "Qwen/Qwen3-Embedding-8B": { + "id": "Qwen/Qwen3-Embedding-8B", + "name": "Qwen3 Embedding 8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2025-12-04", - "last_updated": "2025-12-04", + "release_date": "2025-07-30", + "last_updated": "2025-07-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 160000, - "output": 160000 + "context": 40960, + "output": 4096 + }, + "cost": { + "input": 0.115, + "output": 0.115 } }, - "deepseek/deepseek-v3.1-terminus": { - "id": "deepseek/deepseek-v3.1-terminus", - "name": "DeepSeek/DeepSeek-V3.1-Terminus", + "Qwen/Qwen3-Reranker-4B": { + "id": "Qwen/Qwen3-Reranker-4B", + "name": "Qwen3 Reranker 4B", + "description": "Reranking model for improving retrieval quality in search and recommendation systems", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "tool_call": false, + "release_date": "2025-07-30", + "last_updated": "2025-07-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32000 + "context": 32000, + "output": 4096 + }, + "cost": { + "input": 0.0575, + "output": 0 } }, - "stepfun-ai/gelab-zero-4b-preview": { - "id": "stepfun-ai/gelab-zero-4b-preview", - "name": "Stepfun-Ai/Gelab Zero 4b Preview", + "Qwen/Qwen3.6-35B-A3B-FP8": { + "id": "Qwen/Qwen3.6-35B-A3B-FP8", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 4096 + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.345, + "output": 1.38 } }, - "stepfun/step-3.5-flash": { - "id": "stepfun/step-3.5-flash", - "name": "Stepfun/Step-3.5 Flash", - "attachment": true, + "Qwen/Qwen3-VL-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "name": "Qwen3 VL 30B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "tool_call": true, + "release_date": "2025-07-30", + "last_updated": "2025-07-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 64000, - "output": 4096 + "context": 100000, + "output": 100000 + }, + "cost": { + "input": 0.23, + "output": 0.92 } }, - "x-ai/grok-4-fast": { - "id": "x-ai/grok-4-fast", - "name": "x-AI/Grok-4-Fast", - "attachment": true, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-20", - "last_updated": "2025-09-20", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 65536, + "output": 65536 + }, + "cost": { + "input": 0.23, + "output": 0.92 } }, - "x-ai/grok-code-fast-1": { - "id": "x-ai/grok-code-fast-1", - "name": "x-AI/Grok-Code-Fast 1", + "openai/whisper-large-v3-turbo": { + "id": "openai/whisper-large-v3-turbo", + "name": "Whisper Large v3 Turbo", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-02", - "last_updated": "2025-09-02", + "reasoning": false, + "tool_call": false, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", "modalities": { - "input": ["text"], + "input": ["audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 10000 + "context": 448, + "output": 448 + }, + "cost": { + "input": 0.0023, + "output": 0.0023, + "output_audio": 2.3 } }, - "x-ai/grok-4-fast-reasoning": { - "id": "x-ai/grok-4-fast-reasoning", - "name": "X-Ai/Grok-4-Fast-Reasoning", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "openai/whisper-large-v3": { + "id": "openai/whisper-large-v3", + "name": "Whisper 3 Large", + "description": "Open Whisper checkpoint for robust multilingual transcription and captioning", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 448, + "output": 4096 + }, + "cost": { + "input": 0.0023, + "output": 0.0023, + "output_audio": 2.3 } }, - "x-ai/grok-4.1-fast-non-reasoning": { - "id": "x-ai/grok-4.1-fast-non-reasoning", - "name": "X-Ai/Grok 4.1 Fast Non Reasoning", + "mistralai/Mistral-Medium-3.5-128B": { + "id": "mistralai/Mistral-Medium-3.5-128B", + "name": "Mistral Medium 3.5", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 1.725, + "output": 6.9 } }, - "x-ai/grok-4.1-fast": { - "id": "x-ai/grok-4.1-fast", - "name": "x-AI/Grok-4.1-Fast", + "mistralai/Voxtral-Small-24B-2507": { + "id": "mistralai/Voxtral-Small-24B-2507", + "name": "Voxtral Small 24B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "voxtral", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-11-20", - "last_updated": "2025-11-20", + "reasoning": false, + "tool_call": false, + "release_date": "2025-03-01", + "last_updated": "2025-03-01", "modalities": { - "input": ["text"], + "input": ["audio", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 32000, + "output": 32000 + }, + "cost": { + "input": 0.0023, + "output": 0.0023, + "output_audio": 2.3 } }, - "x-ai/grok-4-fast-non-reasoning": { - "id": "x-ai/grok-4-fast-non-reasoning", - "name": "X-Ai/Grok-4-Fast-Non-Reasoning", + "nvidia/Llama-3.3-70B-Instruct-FP8": { + "id": "nvidia/Llama-3.3-70B-Instruct-FP8", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 1.15, + "output": 1.15 } }, - "x-ai/grok-4.1-fast-reasoning": { - "id": "x-ai/grok-4.1-fast-reasoning", - "name": "X-Ai/Grok 4.1 Fast Reasoning", + "evroc/roc": { + "id": "evroc/roc", + "name": "roc", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "knowledge": "2026-01", + "release_date": "2026-06-06", + "last_updated": "2026-06-06", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 20000000, - "output": 2000000 + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 2.875, + "output": 11.516 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "OpenAI/GPT-5.2", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "KBLab/kb-whisper-large": { + "id": "KBLab/kb-whisper-large", + "name": "KB Whisper", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", "modalities": { - "input": ["text", "image"], + "input": ["audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 448, + "output": 448 + }, + "cost": { + "input": 0.0023, + "output": 0.0023, + "output_audio": 2.3 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "OpenAI/GPT-5", + "intfloat/multilingual-e5-large-instruct": { + "id": "intfloat/multilingual-e5-large-instruct", + "name": "E5 Multi-Lingual Large Embeddings 0.6B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "tool_call": false, + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 512, + "output": 512 + }, + "cost": { + "input": 0.114, + "output": 0.114 } - }, - "z-ai/glm-4.7": { - "id": "z-ai/glm-4.7", - "name": "Z-Ai/GLM 4.7", + } + } + }, + "xiaomi": { + "id": "xiaomi", + "env": ["XIAOMI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.xiaomimimo.com/v1", + "name": "Xiaomi", + "doc": "https://platform.xiaomimimo.com/#/docs", + "models": { + "mimo-v2.5-pro-ultraspeed": { + "id": "mimo-v2.5-pro-ultraspeed", + "name": "MiMo-V2.5-Pro-UltraSpeed", + "description": "MiMo pro model for strong multimodal reasoning and agent execution", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2024-12", + "release_date": "2026-06-08", + "last_updated": "2026-06-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 1048576, + "output": 131072 + }, + "status": "beta", + "cost": { + "input": 1.305, + "output": 2.61, + "cache_read": 0.0108 } }, - "z-ai/glm-5": { - "id": "z-ai/glm-5", - "name": "Z-Ai/GLM 5", - "attachment": false, + "mimo-v2.5": { + "id": "mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-06-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "z-ai/autoglm-phone-9b": { - "id": "z-ai/autoglm-phone-9b", - "name": "Z-Ai/Autoglm Phone 9b", + "mimo-v2-omni": { + "id": "mimo-v2-omni", + "name": "MiMo-V2-Omni", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-06-24", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 12800, - "output": 4096 + "context": 262144, + "output": 131072 + }, + "status": "deprecated", + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "z-ai/glm-4.6": { - "id": "z-ai/glm-4.6", - "name": "Z-AI/GLM 4.6", + "mimo-v2-flash": { + "id": "mimo-v2-flash", + "name": "MiMo-V2-Flash", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2025-10-11", + "knowledge": "2024-12-01", + "release_date": "2025-12-16", + "last_updated": "2026-06-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 262144, + "output": 65536 + }, + "status": "deprecated", + "cost": { + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "Minimax/Minimax-M2", + "mimo-v2-pro": { + "id": "mimo-v2-pro", + "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-06-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 1048576, + "output": 131072 + }, + "status": "deprecated", + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "Minimax/Minimax-M2.1", + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-06-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 204800, - "output": 128000 + "context": 1048576, + "output": 131072 + }, + "cost": { + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } - }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "Minimax/Minimax-M2.5", + } + } + }, + "inception": { + "id": "inception", + "env": ["INCEPTION_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.inceptionlabs.ai/v1/", + "name": "Inception", + "doc": "https://platform.inceptionlabs.ai/docs", + "models": { + "mercury-edit-2": { + "id": "mercury-edit-2", + "name": "Mercury Edit 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": false, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-03-30", + "last_updated": "2026-03-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 128000 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.25, + "output": 0.75, + "cache_read": 0.025 } }, - "minimax/minimax-m2.5-highspeed": { - "id": "minimax/minimax-m2.5-highspeed", - "name": "Minimax/Minimax-M2.5 Highspeed", + "mercury-2": { + "id": "mercury-2", + "name": "Mercury 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "mercury", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-02-14", - "last_updated": "2026-02-14", + "knowledge": "2025-01-01", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 128000 + "context": 128000, + "output": 50000 + }, + "cost": { + "input": 0.25, + "output": 0.75, + "cache_read": 0.025 } - }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Moonshotai/Kimi-K2.5", + } + } + }, + "anthropic": { + "id": "anthropic", + "env": ["ANTHROPIC_API_KEY"], + "npm": "@ai-sdk/anthropic", + "name": "Anthropic", + "doc": "https://docs.anthropic.com/en/docs/about-claude/models", + "models": { + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-01-28", - "last_updated": "2026-01-28", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 - } - }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 0905", - "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-09-08", - "last_updated": "2025-09-08", - "modalities": { - "input": ["text"], - "output": ["text"] + "context": 200000, + "output": 64000 }, - "open_weights": false, - "limit": { - "context": 256000, - "output": 100000 + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "attachment": false, - "reasoning": false, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2025-11-07", - "last_updated": "2025-11-07", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 100000 - } - }, - "meituan/longcat-flash-chat": { - "id": "meituan/longcat-flash-chat", - "name": "Meituan/Longcat-Flash-Chat", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2025-11-05", - "last_updated": "2025-11-05", - "modalities": { - "input": ["text"], - "output": ["text"] + "context": 200000, + "output": 64000 }, - "open_weights": false, - "limit": { - "context": 131072, - "output": 131072 + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "meituan/longcat-flash-lite": { - "id": "meituan/longcat-flash-lite", - "name": "Meituan/Longcat-Flash-Lite", - "attachment": false, - "reasoning": false, + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 320000 + "context": 200000, + "output": 32000 + }, + "status": "deprecated", + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "Mimo-V2-Flash", - "family": "mimo", - "attachment": false, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "Xiaomi/Mimo-V2-Flash", - "family": "mimo", - "attachment": false, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-14", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } - } - } - }, - "kilo": { - "id": "kilo", - "env": ["KILO_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.kilo.ai/api/gateway", - "name": "Kilo Gateway", - "doc": "https://kilo.ai", - "models": { - "rekaai/reka-edge": { - "id": "rekaai/reka-edge", - "name": "Reka Edge", + }, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-20", - "last_updated": "2026-04-11", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-29", + "last_updated": "2026-06-30", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "rekaai/reka-flash-3": { - "id": "rekaai/reka-flash-3", - "name": "Reka Flash 3", - "attachment": false, + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-03-12", - "last_updated": "2026-04-11", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "ai21/jamba-large-1.7": { - "id": "ai21/jamba-large-1.7", - "name": "AI21: Jamba Large 1.7", - "attachment": false, - "reasoning": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-09", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 4096 + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 2, - "output": 8 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "alibaba/tongyi-deepresearch-30b-a3b": { - "id": "alibaba/tongyi-deepresearch-30b-a3b", - "name": "Tongyi DeepResearch 30B A3B", - "attachment": false, + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2026-03-15", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "output": 32000 }, + "status": "deprecated", "cost": { - "input": 0.09, - "output": 0.45 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "inflection/inflection-3-pi": { - "id": "inflection/inflection-3-pi", - "name": "Inflection: Inflection 3 Pi", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-10-11", - "last_updated": "2026-03-15", + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "release_date": "2026-06-07", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "output": 1024 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2.5, - "output": 10 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "inflection/inflection-3-productivity": { - "id": "inflection/inflection-3-productivity", - "name": "Inflection: Inflection 3 Productivity", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-10-11", - "last_updated": "2026-03-15", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "output": 1024 + "context": 200000, + "output": 64000 }, "cost": { - "input": 2.5, - "output": 10 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "liquid/lfm-2-24b-a2b": { - "id": "liquid/lfm-2-24b-a2b", - "name": "LiquidAI: LFM2-24B-A2B", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "knowledge": "2025-05-31", + "release_date": "2026-02-04", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 0.03, - "output": 0.12 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "writer/palmyra-x5": { - "id": "writer/palmyra-x5", - "name": "Writer: Palmyra X5", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1040000, - "output": 8192 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "ibm-granite/granite-4.1-8b": { - "id": "ibm-granite/granite-4.1-8b", - "name": "IBM: Granite 4.1 8B", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-30", - "last_updated": "2026-05-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.1, - "cache_read": 0.05 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } - }, - "ibm-granite/granite-4.0-h-micro": { - "id": "ibm-granite/granite-4.0-h-micro", - "name": "IBM: Granite 4.0 Micro", + } + } + }, + "tencent-coding-plan": { + "id": "tencent-coding-plan", + "env": ["TENCENT_CODING_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.lkeap.cloud.tencent.com/coding/v3", + "name": "Tencent Coding Plan (China)", + "doc": "https://cloud.tencent.com/document/product/1772/128947", + "models": { + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-20", - "last_updated": "2026-03-15", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131000, + "context": 204800, "output": 32768 }, "cost": { - "input": 0.017, - "output": 0.11 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "essentialai/rnj-1-instruct": { - "id": "essentialai/rnj-1-instruct", - "name": "EssentialAI: Rnj 1 Instruct", - "attachment": false, - "reasoning": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-12-05", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 6554 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "perplexity/sonar-pro": { - "id": "perplexity/sonar-pro", - "name": "Perplexity: Sonar Pro", - "attachment": true, + "hunyuan-turbos": { + "id": "hunyuan-turbos", + "name": "Hunyuan-TurboS", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "release_date": "2026-03-08", + "last_updated": "2026-03-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 3, - "output": 15 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "perplexity/sonar-deep-research": { - "id": "perplexity/sonar-deep-research", - "name": "Perplexity: Sonar Deep Research", + "hunyuan-t1": { + "id": "hunyuan-t1", + "name": "Hunyuan-T1", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "release_date": "2026-03-08", + "last_updated": "2026-03-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 25600 + "context": 131072, + "output": 16384 }, "cost": { - "input": 2, - "output": 8 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "perplexity/sonar": { - "id": "perplexity/sonar", - "name": "Perplexity: Sonar", - "attachment": true, + "tc-code-latest": { + "id": "tc-code-latest", + "name": "Auto", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "auto", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "release_date": "2026-03-08", + "last_updated": "2026-03-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 127072, - "output": 25415 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1, - "output": 1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "perplexity/sonar-pro-search": { - "id": "perplexity/sonar-pro-search", - "name": "Perplexity: Sonar Pro Search", - "attachment": true, + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-31", - "last_updated": "2026-03-15", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 3, - "output": 15 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "perplexity/sonar-reasoning-pro": { - "id": "perplexity/sonar-reasoning-pro", - "name": "Perplexity: Sonar Reasoning Pro", - "attachment": true, - "reasoning": true, - "tool_call": false, + "hunyuan-2.0-instruct": { + "id": "hunyuan-2.0-instruct", + "name": "Tencent HY 2.0 Instruct", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", + "attachment": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "release_date": "2026-03-08", + "last_updated": "2026-03-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 25600 + "context": 131072, + "output": 16384 }, "cost": { - "input": 2, - "output": 8 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek/deepseek-chat-v3.1": { - "id": "deepseek/deepseek-chat-v3.1", - "name": "DeepSeek: DeepSeek V3.1", + "hunyuan-2.0-thinking": { + "id": "hunyuan-2.0-thinking", + "name": "Tencent HY 2.0 Think", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "release_date": "2026-03-08", + "last_updated": "2026-03-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 7168 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "deepseek/deepseek-chat": { - "id": "deepseek/deepseek-chat", - "name": "DeepSeek: DeepSeek V3", - "attachment": false, - "reasoning": false, + } + } + }, + "freemodel": { + "id": "freemodel", + "env": ["FREEMODEL_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://cc.freemodel.dev/v1", + "name": "FreeModel", + "doc": "https://freemodel.dev", + "models": { + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2026-03-15", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.32, - "output": 0.89, - "cache_read": 0.15 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "deepseek/deepseek-r1-distill-llama-70b": { - "id": "deepseek/deepseek-r1-distill-llama-70b", - "name": "DeepSeek: R1 Distill Llama 70B", - "attachment": false, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-01-23", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.7, - "output": 0.8, - "cache_read": 0.015 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepseek/deepseek-r1": { - "id": "deepseek/deepseek-r1", - "name": "DeepSeek: R1", - "attachment": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 16000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.7, - "output": 2.5 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepseek/deepseek-v3.2-speciale": { - "id": "deepseek/deepseek-v3.2-speciale", - "name": "DeepSeek: DeepSeek V3.2 Speciale", - "attachment": false, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.freemodel.dev/v1" }, "cost": { - "input": 0.4, - "output": 1.2, - "cache_read": 0.135 + "input": 1.75, + "output": 14, + "cache_read": 0.175, + "cache_write": 1.75 } }, - "deepseek/deepseek-r1-distill-qwen-32b": { - "id": "deepseek/deepseek-r1-distill-qwen-32b", - "name": "DeepSeek: R1 Distill Qwen 32B", - "attachment": false, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-11-25", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.29, - "output": 0.29 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "deepseek/deepseek-v3.2-exp": { - "id": "deepseek/deepseek-v3.2-exp", - "name": "DeepSeek: DeepSeek V3.2 Exp", - "attachment": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-09-29", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.freemodel.dev/v1" }, "cost": { - "input": 0.27, - "output": 0.41 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 2.5 } }, - "deepseek/deepseek-v4-flash": { - "id": "deepseek/deepseek-v4-flash", - "name": "DeepSeek: DeepSeek V4 Flash", - "attachment": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 384000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.freemodel.dev/v1" }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.0028 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075, + "cache_write": 0.75 } }, - "deepseek/deepseek-v4-pro": { - "id": "deepseek/deepseek-v4-pro", - "name": "DeepSeek: DeepSeek V4 Pro", - "attachment": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-05-01", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 384000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.435, - "output": 0.87, - "cache_read": 0.003625 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "DeepSeek: DeepSeek V3.2", - "attachment": false, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2026-03-15", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.26, - "output": 0.38, - "cache_read": 0.125 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "deepseek/deepseek-chat-v3-0324": { - "id": "deepseek/deepseek-chat-v3-0324", - "name": "DeepSeek: DeepSeek V3 0324", - "attachment": false, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-03-24", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.freemodel.dev/v1" }, "cost": { - "input": 0.2, - "output": 0.77, - "cache_read": 0.095 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 5 } - }, - "deepseek/deepseek-r1-0528": { - "id": "deepseek/deepseek-r1-0528", - "name": "DeepSeek: R1 0528", - "attachment": false, + } + } + }, + "sap-ai-core": { + "id": "sap-ai-core", + "env": ["AICORE_SERVICE_KEY"], + "npm": "@jerome-benoit/sap-ai-provider-v2", + "name": "SAP AI Core", + "doc": "https://help.sap.com/docs/sap-ai-core", + "models": { + "anthropic--claude-4.8-opus": { + "id": "anthropic--claude-4.8-opus", + "name": "anthropic--claude-4.8-opus", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 2.15, - "cache_read": 0.2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepseek/deepseek-v3.1-terminus": { - "id": "deepseek/deepseek-v3.1-terminus", - "name": "DeepSeek: DeepSeek V3.1 Terminus", - "attachment": false, + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "gemini-3.1-flash-lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.21, - "output": 0.79, - "cache_read": 0.13 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 } }, - "openrouter/auto": { - "id": "openrouter/auto", - "name": "Auto Router", + "anthropic--claude-4.6-sonnet": { + "id": "anthropic--claude-4.6-sonnet", + "name": "anthropic--claude-4.6-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "knowledge": "2025-08", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], - "output": ["image", "text"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openrouter/bodybuilder": { - "id": "openrouter/bodybuilder", - "name": "Body Builder (beta)", - "attachment": false, + "anthropic--claude-3-sonnet": { + "id": "anthropic--claude-3-sonnet", + "name": "anthropic--claude-3-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": false, - "tool_call": false, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": true, + "knowledge": "2023-08-31", + "release_date": "2024-03-04", + "last_updated": "2024-03-04", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 200000, + "output": 4096 }, - "status": "beta", "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openrouter/owl-alpha": { - "id": "openrouter/owl-alpha", - "name": "Owl Alpha", - "attachment": false, + "anthropic--claude-4-sonnet": { + "id": "anthropic--claude-4-sonnet", + "name": "anthropic--claude-4-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-30", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1048756, - "output": 262144 - }, - "status": "alpha", - "cost": { - "input": 0, - "output": 0 - } - }, - "openrouter/pareto-code": { - "id": "openrouter/pareto-code", - "name": "Pareto Code Router", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-04-21", - "last_updated": "2026-05-01", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 65536 + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openrouter/free": { - "id": "openrouter/free", - "name": "Free Models Router", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "gemini-2.5-pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-03-25", + "last_updated": "2025-06-05", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "inclusionai/ling-2.6-1t:free": { - "id": "inclusionai/ling-2.6-1t:free", - "name": "inclusionAI: Ling-2.6-1T (free)", - "attachment": false, - "reasoning": false, + "gpt-5": { + "id": "gpt-5", + "name": "gpt-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-23", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "inclusionai/ling-2.6-flash": { - "id": "inclusionai/ling-2.6-flash", - "name": "inclusionAI: Ling-2.6 Flash", - "attachment": false, - "reasoning": false, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "gemini-2.5-flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-05-01", + "knowledge": "2025-01", + "release_date": "2025-04-17", + "last_updated": "2025-06-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.08, - "output": 0.24, - "cache_read": 0.016 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 } }, - "arcee-ai/trinity-mini": { - "id": "arcee-ai/trinity-mini", - "name": "Arcee AI: Trinity Mini", - "attachment": false, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "gemini-3.5-flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-12", - "last_updated": "2026-01-28", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.045, - "output": 0.15 + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 } }, - "arcee-ai/virtuoso-large": { - "id": "arcee-ai/virtuoso-large", - "name": "Arcee AI: Virtuoso Large", - "attachment": false, - "reasoning": false, + "anthropic--claude-4.5-haiku": { + "id": "anthropic--claude-4.5-haiku", + "name": "anthropic--claude-4.5-haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-05-06", - "last_updated": "2026-03-15", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 200000, "output": 64000 }, "cost": { - "input": 0.75, - "output": 1.2 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "arcee-ai/trinity-large-thinking": { - "id": "arcee-ai/trinity-large-thinking", - "name": "Arcee AI: Trinity Large Thinking", - "attachment": false, - "reasoning": true, + "anthropic--claude-3-haiku": { + "id": "anthropic--claude-3-haiku", + "name": "anthropic--claude-3-haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-11", + "knowledge": "2023-08-31", + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "output": 4096 }, "cost": { - "input": 0.22, - "output": 0.85 + "input": 0.25, + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 } }, - "arcee-ai/spotlight": { - "id": "arcee-ai/spotlight", - "name": "Arcee AI: Spotlight", + "anthropic--claude-4-opus": { + "id": "anthropic--claude-4-opus", + "name": "anthropic--claude-4-opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-05-06", - "last_updated": "2026-03-15", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65537 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "arcee-ai/maestro-reasoning": { - "id": "arcee-ai/maestro-reasoning", - "name": "Arcee AI: Maestro Reasoning", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic--claude-4.5-sonnet": { + "id": "anthropic--claude-4.5-sonnet", + "name": "anthropic--claude-4.5-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-05-06", - "last_updated": "2026-03-15", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.9, - "output": 3.3 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "arcee-ai/coder-large": { - "id": "arcee-ai/coder-large", - "name": "Arcee AI: Coder Large", - "attachment": false, + "anthropic--claude-3.5-sonnet": { + "id": "anthropic--claude-3.5-sonnet", + "name": "anthropic--claude-3.5-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-05-06", - "last_updated": "2026-03-15", + "knowledge": "2024-04-30", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 0.8 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "arcee-ai/trinity-large-preview": { - "id": "arcee-ai/trinity-large-preview", - "name": "Arcee AI: Trinity Large Preview", - "attachment": false, - "reasoning": false, + "anthropic--claude-4.6-opus": { + "id": "anthropic--claude-4.6-opus", + "name": "anthropic--claude-4.6-opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-01-28", - "last_updated": "2026-05-01", + "knowledge": "2025-05", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.45 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "deepcogito/cogito-v2.1-671b": { - "id": "deepcogito/cogito-v2.1-671b", - "name": "Deep Cogito: Cogito v2.1 671B", - "attachment": false, + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "gemini-2.5-flash-lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.25, - "output": 1.25 + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 } }, - "upstage/solar-pro-3": { - "id": "upstage/solar-pro-3", - "name": "Upstage: Solar Pro 3", - "attachment": false, + "anthropic--claude-3.7-sonnet": { + "id": "anthropic--claude-3.7-sonnet", + "name": "anthropic--claude-3.7-sonnet", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-03-15", + "knowledge": "2024-10-31", + "release_date": "2025-02-24", + "last_updated": "2025-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "nex-agi/deepseek-v3.1-nex-n1": { - "id": "nex-agi/deepseek-v3.1-nex-n1", - "name": "Nex AGI: DeepSeek V3.1 Nex N1", + "sonar": { + "id": "sonar", + "name": "sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-11-25", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 163840 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.27, + "input": 1, "output": 1 } }, - "bytedance-seed/seed-1.6": { - "id": "bytedance-seed/seed-1.6", - "name": "ByteDance Seed: Seed 1.6", + "sonar-pro": { + "id": "sonar-pro", + "name": "sonar-pro", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "family": "sonar-pro", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-09", - "last_updated": "2025-09", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 2 + "input": 3, + "output": 15 } }, - "bytedance-seed/seed-2.0-lite": { - "id": "bytedance-seed/seed-2.0-lite", - "name": "ByteDance Seed: Seed-2.0-Lite", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "gpt-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-10", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 2 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "bytedance-seed/seed-1.6-flash": { - "id": "bytedance-seed/seed-1.6-flash", - "name": "ByteDance Seed: Seed 1.6 Flash", + "gpt-4.1": { + "id": "gpt-4.1", + "name": "gpt-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2026-03-15", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 1047576, "output": 32768 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "bytedance-seed/seed-2.0-mini": { - "id": "bytedance-seed/seed-2.0-mini", - "name": "ByteDance Seed: Seed-2.0-Mini", + "anthropic--claude-3-opus": { + "id": "anthropic--claude-3-opus", + "name": "anthropic--claude-3-opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-02-27", - "last_updated": "2026-03-15", + "knowledge": "2023-08-31", + "release_date": "2024-02-29", + "last_updated": "2024-02-29", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 200000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "mancer/weaver": { - "id": "mancer/weaver", - "name": "Mancer: Weaver (alpha)", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2023-08-02", - "last_updated": "2026-03-15", + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "gpt-5-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "output": 2000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.75, - "output": 1 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "anthracite-org/magnum-v4-72b": { - "id": "anthracite-org/magnum-v4-72b", - "name": "Magnum v4 72B", - "attachment": false, + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "gpt-4.1-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-10-22", - "last_updated": "2026-03-15", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, - "output": 2048 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 3, - "output": 5 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "~google/gemini-pro-latest": { - "id": "~google/gemini-pro-latest", - "name": "Google: Gemini Pro Latest", - "attachment": true, + "sonar-deep-research": { + "id": "sonar-deep-research", + "name": "sonar-deep-research", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar-deep-research", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-02-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "output": 32768 }, "cost": { "input": 2, - "output": 12, - "cache_read": 0.2, - "cache_write": 0.375 + "output": 8, + "reasoning": 3 } }, - "~google/gemini-flash-latest": { - "id": "~google/gemini-flash-latest", - "name": "Google: Gemini Flash Latest", + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "gpt-5-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.08333333333333334 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "kilo-auto/balanced": { - "id": "kilo-auto/balanced", - "name": "Kilo Auto Balanced", - "attachment": false, + "anthropic--claude-4.5-opus": { + "id": "anthropic--claude-4.5-opus", + "name": "anthropic--claude-4.5-opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 3 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "kilo-auto/frontier": { - "id": "kilo-auto/frontier", - "name": "Kilo Auto Frontier", + "anthropic--claude-4.7-opus": { + "id": "anthropic--claude-4.7-opus", + "name": "anthropic--claude-4.7-opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, @@ -46561,922 +90277,1451 @@ }, "cost": { "input": 5, - "output": 25 + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "kilo-auto/small": { - "id": "kilo-auto/small", - "name": "Kilo Auto Small", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "gpt-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.4 + "input": 5, + "output": 30, + "cache_read": 0.5 } - }, - "kilo-auto/free": { - "id": "kilo-auto/free", - "name": "Kilo Auto Free", + } + } + }, + "opencode": { + "id": "opencode", + "env": ["OPENCODE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://opencode.ai/zen/v1", + "name": "OpenCode Zen", + "doc": "https://opencode.ai/docs/zen", + "models": { + "ring-2.6-1t-free": { + "id": "ring-2.6-1t-free", + "name": "Ring 2.6 1T Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "ring-1t-free", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "knowledge": "2025-06", + "release_date": "2026-05-08", + "last_updated": "2026-05-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262000, + "output": 66000 }, + "status": "deprecated", "cost": { "input": 0, "output": 0 } }, - "undi95/remm-slerp-l2-13b": { - "id": "undi95/remm-slerp-l2-13b", - "name": "ReMM SLERP 13B", - "attachment": false, - "reasoning": false, - "tool_call": false, + "mimo-v2-pro-free": { + "id": "mimo-v2-pro-free", + "name": "MiMo V2 Pro Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo-pro-free", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2023-07-22", - "last_updated": "2026-03-15", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 6144, - "output": 4096 + "context": 1048576, + "output": 64000 }, + "status": "deprecated", "cost": { - "input": 0.45, - "output": 0.65 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "allenai/olmo-3-32b-think": { - "id": "allenai/olmo-3-32b-think", - "name": "AllenAI: Olmo 3 32B Think", + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-11-22", - "last_updated": "2026-03-15", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.15, - "output": 0.5 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "nousresearch/hermes-2-pro-llama-3-8b": { - "id": "nousresearch/hermes-2-pro-llama-3-8b", - "name": "NousResearch: Hermes 2 Pro - Llama-3 8B", + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-05-27", - "last_updated": "2024-06-27", + "knowledge": "2025-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.14 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "nousresearch/hermes-4-405b": { - "id": "nousresearch/hermes-4-405b", - "name": "Nous: Hermes 4 405B", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-08-25", - "last_updated": "2025-08-25", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 204800, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 1, - "output": 3 + "input": 0.6, + "output": 2.2, + "cache_read": 0.1 } }, - "nousresearch/hermes-3-llama-3.1-70b": { - "id": "nousresearch/hermes-3-llama-3.1-70b", - "name": "Nous: Hermes 3 70B Instruct", - "attachment": false, - "reasoning": false, - "tool_call": false, + "mimo-v2.5-free": { + "id": "mimo-v2.5-free", + "name": "MiMo V2.5 Free", + "description": "MiMo omni model for text, image, video, audio, and agents", + "family": "mimo-v2.5-free", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-18", - "last_updated": "2026-03-15", + "knowledge": "2024-12", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "nousresearch/hermes-4-70b": { - "id": "nousresearch/hermes-4-70b", - "name": "Nous: Hermes 4 70B", + "kimi-k2": { + "id": "kimi-k2", + "name": "Kimi K2", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2025-08-25", - "last_updated": "2026-03-15", + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.13, - "output": 0.4, - "cache_read": 0.055 + "input": 0.4, + "output": 2.5, + "cache_read": 0.4 } }, - "nousresearch/hermes-3-llama-3.1-405b": { - "id": "nousresearch/hermes-3-llama-3.1-405b", - "name": "Nous: Hermes 3 405B Instruct", + "minimax-m2.1": { + "id": "minimax-m2.1", + "name": "MiniMax-M2.1", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-16", - "last_updated": "2024-08-16", + "knowledge": "2025-01", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 204800, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 1, - "output": 1 + "input": 0.3, + "output": 1.2, + "cache_read": 0.1 } }, - "morph/morph-v3-fast": { - "id": "morph/morph-v3-fast", - "name": "Morph: Morph V3 Fast", + "nemotron-3-ultra-free": { + "id": "nemotron-3-ultra-free", + "name": "Nemotron 3 Ultra Free", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "knowledge": "2026-02", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 81920, - "output": 38000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.8, - "output": 1.2 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "morph/morph-v3-large": { - "id": "morph/morph-v3-large", - "name": "Morph: Morph V3 Large", + "glm-4.7-free": { + "id": "glm-4.7-free", + "name": "GLM-4.7 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, + "context": 204800, "output": 131072 }, + "status": "deprecated", "cost": { - "input": 0.9, - "output": 1.9 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "stepfun/step-3.5-flash:free": { - "id": "stepfun/step-3.5-flash:free", - "name": "StepFun: Step 3.5 Flash (free)", - "attachment": false, + "gemini-3-flash": { + "id": "gemini-3-flash", + "name": "Gemini 3 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-26", - "last_updated": "2026-05-01", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/google" }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "stepfun/step-3.5-flash": { - "id": "stepfun/step-3.5-flash", - "name": "StepFun: Step 3.5 Flash", + "deepseek-v4-flash-free": { + "id": "deepseek-v4-flash-free", + "name": "DeepSeek V4 Flash Free", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek-flash-free", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-01-29", - "last_updated": "2026-01-29", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.02 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "alpindale/goliath-120b": { - "id": "alpindale/goliath-120b", - "name": "Goliath 120B", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-sonnet-4": { + "id": "claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2023-11-10", - "last_updated": "2026-03-15", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 6144, - "output": 1024 + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 3.75, - "output": 7.5 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "mistralai/mistral-nemo": { - "id": "mistralai/mistral-nemo", - "name": "Mistral: Mistral Nemo", - "attachment": false, - "reasoning": false, + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-07-01", - "last_updated": "2024-07-30", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.02, - "output": 0.04 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "mistralai/mistral-saba": { - "id": "mistralai/mistral-saba", - "name": "Mistral: Saba", - "attachment": false, - "reasoning": false, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-02-17", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 1.07, + "output": 8.5, + "cache_read": 0.107 } }, - "mistralai/mistral-large-2512": { - "id": "mistralai/mistral-large-2512", - "name": "Mistral: Mistral Large 3 2512", + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-11-01", - "last_updated": "2025-12-16", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 52429 + "context": 1048576, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/google" }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 } }, - "mistralai/devstral-medium": { - "id": "mistralai/devstral-medium", - "name": "Mistral: Devstral Medium", + "minimax-m3-free": { + "id": "minimax-m3-free", + "name": "MiniMax-M3 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax-m3-free", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "knowledge": "2025-01", + "release_date": "2026-05-31", + "last_updated": "2026-05-31", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 200000, + "output": 32000 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.4, - "output": 2 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "mistralai/mistral-small-3.1-24b-instruct": { - "id": "mistralai/mistral-small-3.1-24b-instruct", - "name": "Mistral: Mistral Small 3.1 24B", + "gemini-3-pro": { + "id": "gemini-3-pro", + "name": "Gemini 3 Pro", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gemini-pro", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-03-17", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 131072 + "context": 1048576, + "output": 65536 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/google" }, "cost": { - "input": 0.35, - "output": 0.56, - "cache_read": 0.015 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "mistralai/mistral-medium-3-5": { - "id": "mistralai/mistral-medium-3-5", - "name": "Mistral: Mistral Medium 3.5", + "kimi-k2.5-free": { + "id": "kimi-k2.5-free", + "name": "Kimi K2.5 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-free", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-04-30", - "last_updated": "2026-05-07", + "knowledge": "2024-10", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 262144, "output": 262144 }, + "status": "deprecated", "cost": { - "input": 1.5, - "output": 7.5 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "mistralai/pixtral-large-2411": { - "id": "mistralai/pixtral-large-2411", - "name": "Mistral: Pixtral Large 2411", + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2024-11-19", - "last_updated": "2026-03-15", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 2, - "output": 6 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "mistralai/devstral-2512": { - "id": "mistralai/devstral-2512", - "name": "Mistral: Devstral 2 2512", + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-09-12", - "last_updated": "2026-03-15", + "knowledge": "2025-04", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.025 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "mistralai/codestral-2508": { - "id": "mistralai/codestral-2508", - "name": "Mistral: Codestral 2508", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 51200 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 1.74, + "output": 3.84, + "cache_read": 0.145 } }, - "mistralai/mistral-small-24b-instruct-2501": { - "id": "mistralai/mistral-small-24b-instruct-2501", - "name": "Mistral: Mistral Small 3", + "glm-4.6": { + "id": "glm-4.6", + "name": "GLM-4.6", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-12-29", - "last_updated": "2026-01-10", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 16384 + "context": 204800, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 0.05, - "output": 0.08 + "input": 0.6, + "output": 2.2, + "cache_read": 0.1 } }, - "mistralai/mistral-large-2411": { - "id": "mistralai/mistral-large-2411", - "name": "Mistral Large 2411", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-07-24", - "last_updated": "2024-11-04", + "knowledge": "2024-10", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 2, - "output": 6 + "input": 0.4, + "output": 2.5, + "cache_read": 0.4 } }, - "mistralai/mixtral-8x22b-instruct": { - "id": "mistralai/mixtral-8x22b-instruct", - "name": "Mistral: Mixtral 8x22B Instruct", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "interleaved": true, "temperature": true, - "release_date": "2024-04-17", - "last_updated": "2024-04-17", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 13108 + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 2, - "output": 6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "mistralai/mistral-large-2407": { - "id": "mistralai/mistral-large-2407", - "name": "Mistral Large 2407", - "attachment": false, - "reasoning": false, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-11-19", - "last_updated": "2026-03-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 2, - "output": 6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "mistralai/ministral-8b-2512": { - "id": "mistralai/ministral-8b-2512", - "name": "Mistral: Ministral 3 8B 2512", + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2026-03-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "mistralai/mistral-medium-3.1": { - "id": "mistralai/mistral-medium-3.1", - "name": "Mistral: Mistral Medium 3.1", + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "mistralai/mistral-small-2603": { - "id": "mistralai/mistral-small-2603", - "name": "Mistral: Mistral Small 4", - "attachment": true, - "reasoning": true, + "qwen3-coder": { + "id": "qwen3-coder", + "name": "Qwen3 Coder", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-04-11", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 65536 }, + "status": "deprecated", "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.015 + "input": 0.45, + "output": 1.8 } }, - "mistralai/ministral-3b-2512": { - "id": "mistralai/ministral-3b-2512", - "name": "Mistral: Ministral 3 3B 2512", - "attachment": true, - "reasoning": false, + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2026-03-15", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "mistralai/voxtral-small-24b-2507": { - "id": "mistralai/voxtral-small-24b-2507", - "name": "Mistral: Voxtral Small 24B 2507", + "hy3-preview-free": { + "id": "hy3-preview-free", + "name": "Hy3 preview Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "hy3-free", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "knowledge": "2025-06", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 6400 + "context": 256000, + "output": 64000 }, + "status": "deprecated", "cost": { - "input": 0.1, - "output": 0.3 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "mistralai/mixtral-8x7b-instruct": { - "id": "mistralai/mixtral-8x7b-instruct", - "name": "Mistral: Mixtral 8x7B Instruct", - "attachment": false, - "reasoning": false, + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2023-12-10", - "last_updated": "2026-03-15", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 16384 + "context": 512000, + "output": 128000 }, "cost": { - "input": 0.54, - "output": 0.54 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "mistralai/mistral-medium-3": { - "id": "mistralai/mistral-medium-3", - "name": "Mistral: Mistral Medium 3", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.4, - "output": 2 + "input": 1.07, + "output": 8.5, + "cache_read": 0.107 } }, - "mistralai/mistral-small-3.2-24b-instruct": { - "id": "mistralai/mistral-small-3.2-24b-instruct", - "name": "Mistral: Mistral Small 3.2 24B", - "attachment": true, - "reasoning": false, + "gpt-5.3-codex-spark": { + "id": "gpt-5.3-codex-spark", + "name": "GPT-5.3 Codex Spark", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex-spark", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-06-20", - "last_updated": "2025-06-20", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.06, - "output": 0.18, - "cache_read": 0.03 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "mistralai/devstral-small": { - "id": "mistralai/devstral-small", - "name": "Mistral: Devstral Small 1.1", - "attachment": false, - "reasoning": false, + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-05-07", - "last_updated": "2025-07-10", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "mistralai/mistral-large": { - "id": "mistralai/mistral-large", - "name": "Mistral Large", - "attachment": false, - "reasoning": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-07-24", - "last_updated": "2025-12-02", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 25600 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 2, - "output": 6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "mistralai/mistral-7b-instruct-v0.1": { - "id": "mistralai/mistral-7b-instruct-v0.1", - "name": "Mistral: Mistral 7B Instruct v0.1", + "ling-2.6-flash-free": { + "id": "ling-2.6-flash-free", + "name": "Ling 2.6 Flash Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "ling-flash-free", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-06", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2824, - "output": 565 + "context": 262100, + "output": 32800 }, + "status": "deprecated", "cost": { - "input": 0.11, - "output": 0.19 + "input": 0, + "output": 0 } }, - "mistralai/ministral-14b-2512": { - "id": "mistralai/ministral-14b-2512", - "name": "Mistral: Ministral 3 14B 2512", + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.5", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 262144, - "output": 52429 + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { "input": 0.2, - "output": 0.2 + "output": 1.2, + "cache_read": 0.02, + "cache_write": 0.25 } }, - "~anthropic/claude-haiku-latest": { - "id": "~anthropic/claude-haiku-latest", - "name": "Anthropic: Claude Haiku Latest", + "claude-3-5-haiku": { + "id": "claude-3-5-haiku", + "name": "Claude Haiku 3.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-haiku", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -47484,1148 +91729,1840 @@ "open_weights": false, "limit": { "context": 200000, - "output": 64000 + "output": 8192 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "~anthropic/claude-sonnet-latest": { - "id": "~anthropic/claude-sonnet-latest", - "name": "Anthropic: Claude Sonnet Latest", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, + "input": 272000, "output": 128000 }, + "provider": { + "npm": "@ai-sdk/openai" + }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "~anthropic/claude-opus-latest": { - "id": "~anthropic/claude-opus-latest", - "name": "Anthropic: Claude Opus Latest", + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2026-04-16", - "last_updated": "2026-05-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, + "input": 272000, "output": 128000 }, + "provider": { + "npm": "@ai-sdk/openai" + }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "meta-llama/llama-3.3-70b-instruct": { - "id": "meta-llama/llama-3.3-70b-instruct", - "name": "Meta: Llama 3.3 70B Instruct", + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-01", - "last_updated": "2026-02-04", + "knowledge": "2025-01", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.32 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "meta-llama/llama-4-scout": { - "id": "meta-llama/llama-4-scout", - "name": "Meta: Llama 4 Scout", + "grok-4.5": { + "id": "grok-4.5", + "name": "Grok 4.5", + "description": "xAI's latest Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 327680, - "output": 16384 + "context": 500000, + "output": 500000 }, "cost": { - "input": 0.08, - "output": 0.3 + "input": 2, + "output": 6, + "cache_read": 0.5, + "tiers": [ + { + "input": 4, + "output": 12, + "cache_read": 1, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 12, + "cache_read": 1 + } } }, - "meta-llama/llama-guard-3-8b": { - "id": "meta-llama/llama-guard-3-8b", - "name": "Llama Guard 3 8B", + "north-mini-code-free": { + "id": "north-mini-code-free", + "name": "North Mini Code Free", + "description": "Cohere coding model for practical software engineering and agentic edits", + "family": "north-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-04-18", - "last_updated": "2026-02-04", + "knowledge": "2025-09-23", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.02, - "output": 0.06 + "input": 0, + "output": 0 } }, - "meta-llama/llama-4-maverick": { - "id": "meta-llama/llama-4-maverick", - "name": "Meta: Llama 4 Maverick", + "grok-code": { + "id": "grok-code", + "name": "Grok Code Fast 1", + "description": "Legacy model retained for compatibility with older integrations", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-12-24", + "release_date": "2025-08-20", + "last_updated": "2025-08-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 16384 + "context": 256000, + "output": 256000 }, + "status": "deprecated", "cost": { - "input": 0.15, - "output": 0.6 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "meta-llama/llama-3.2-11b-vision-instruct": { - "id": "meta-llama/llama-3.2-11b-vision-instruct", - "name": "Meta: Llama 3.2 11B Vision Instruct", + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.049, - "output": 0.049 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25, + "tiers": [ + { + "input": 2, + "output": 9, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 9, + "cache_read": 0.2, + "cache_write": 2.5 + } } }, - "meta-llama/llama-guard-4-12b": { - "id": "meta-llama/llama-guard-4-12b", - "name": "Meta: Llama Guard 4 12B", + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 + "context": 200000, + "output": 32000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "meta-llama/llama-3.1-70b-instruct": { - "id": "meta-llama/llama-3.1-70b-instruct", - "name": "Meta: Llama 3.1 70B Instruct", - "attachment": false, - "reasoning": false, + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-07-16", - "last_updated": "2024-07-23", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.025 + } + }, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.4, - "output": 0.4 - } - }, - "meta-llama/llama-3.2-1b-instruct": { - "id": "meta-llama/llama-3.2-1b-instruct", - "name": "Meta: Llama 3.2 1B Instruct", - "attachment": false, - "reasoning": false, - "tool_call": false, "temperature": true, - "release_date": "2024-09-18", + "knowledge": "2024-10", + "release_date": "2026-01-27", "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 60000, - "output": 12000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.027, - "output": 0.2 + "input": 0.6, + "output": 3, + "cache_read": 0.08 } }, - "meta-llama/llama-3.2-3b-instruct": { - "id": "meta-llama/llama-3.2-3b-instruct", - "name": "Meta: Llama 3.2 3B Instruct", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2026-03-15", + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 80000, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.051, - "output": 0.34 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 6.25, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "cache_write": 6.25 + } } }, - "meta-llama/llama-3-8b-instruct": { - "id": "meta-llama/llama-3-8b-instruct", - "name": "Meta: Llama 3 8B Instruct", - "attachment": false, - "reasoning": false, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-04-25", - "last_updated": "2025-04-03", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 16384 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.03, - "output": 0.04 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "meta-llama/llama-3.1-8b-instruct": { - "id": "meta-llama/llama-3.1-8b-instruct", - "name": "Meta: Llama 3.1 8B Instruct", + "hy3-free": { + "id": "hy3-free", + "name": "Hy3 Free", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hy3-free", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2025-12-23", + "release_date": "2026-06-26", + "last_updated": "2026-06-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 + "context": 190000, + "output": 64000 }, "cost": { - "input": 0.02, - "output": 0.05 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "meta-llama/llama-3-70b-instruct": { - "id": "meta-llama/llama-3-70b-instruct", - "name": "Meta: Llama 3 70B Instruct", - "attachment": false, - "reasoning": false, - "tool_call": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8000 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.51, - "output": 0.74 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "x-ai/grok-4.20": { - "id": "x-ai/grok-4.20", - "name": "xAI: Grok 4.20", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-31", - "last_updated": "2026-04-11", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "x-ai/grok-code-fast-1:optimized:free": { - "id": "x-ai/grok-code-fast-1:optimized:free", - "name": "xAI: Grok Code Fast 1 Optimized (experimental, free)", - "attachment": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-27", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "x-ai/grok-4.3": { - "id": "x-ai/grok-4.3", - "name": "xAI: Grok 4.3", - "attachment": true, + "minimax-m2.5-free": { + "id": "minimax-m2.5-free", + "name": "MiniMax-M2.5 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax-free", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-05-01", - "last_updated": "2026-05-01", + "knowledge": "2025-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 4096 + "context": 204800, + "output": 131072 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "x-ai/grok-4-fast": { - "id": "x-ai/grok-4-fast", - "name": "xAI: Grok 4 Fast", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "knowledge": "2024-10", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "x-ai/grok-code-fast-1": { - "id": "x-ai/grok-code-fast-1", - "name": "xAI: Grok Code Fast 1", - "attachment": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "x-ai/grok-3-beta": { - "id": "x-ai/grok-3-beta", - "name": "xAI: Grok 3 Beta", + "minimax-m2.1-free": { + "id": "minimax-m2.1-free", + "name": "MiniMax-M2.1 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax-free", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2025-01", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 204800, + "output": 131072 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "x-ai/grok-4": { - "id": "x-ai/grok-4", - "name": "xAI: Grok 4", + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 51200 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "x-ai/grok-3-mini": { - "id": "x-ai/grok-3-mini", - "name": "xAI: Grok 3 Mini", - "attachment": false, + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075 + "input": 30, + "output": 180, + "cache_read": 30 } }, - "x-ai/grok-4.1-fast": { - "id": "x-ai/grok-4.1-fast", - "name": "xAI: Grok 4.1 Fast", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "interleaved": true, "temperature": true, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 1000000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "x-ai/grok-3-mini-beta": { - "id": "x-ai/grok-3-mini-beta", - "name": "xAI: Grok 3 Mini Beta", + "mimo-v2-flash-free": { + "id": "mimo-v2-flash-free", + "name": "MiMo V2 Flash Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo-flash-free", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2024-12", + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 262144, + "output": 65536 }, + "status": "deprecated", "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "x-ai/grok-4.20-multi-agent": { - "id": "x-ai/grok-4.20-multi-agent", - "name": "xAI: Grok 4.20 Multi-Agent", + "gemini-3.1-pro": { + "id": "gemini-3.1-pro", + "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-31", - "last_updated": "2026-04-11", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1048576, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/google" }, "cost": { "input": 2, - "output": 6, - "cache_read": 0.2 + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "x-ai/grok-3": { - "id": "x-ai/grok-3", - "name": "xAI: Grok 3", + "trinity-large-preview-free": { + "id": "trinity-large-preview-free", + "name": "Trinity Large Preview", + "description": "Legacy model retained for compatibility with older integrations", + "family": "trinity", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2025-06", + "release_date": "2026-01-28", + "last_updated": "2026-01-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 26215 + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0, + "output": 0 } }, - "tencent/hy3-preview:free": { - "id": "tencent/hy3-preview:free", - "name": "Tencent: Hy3 Preview (free)", + "big-pickle": { + "id": "big-pickle", + "name": "Big Pickle", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "big-pickle", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-05-01", + "knowledge": "2025-01", + "release_date": "2025-10-17", + "last_updated": "2025-10-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "input": 160000, + "output": 32000 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "tencent/hunyuan-a13b-instruct": { - "id": "tencent/hunyuan-a13b-instruct", - "name": "Tencent: Hunyuan A13B Instruct", - "attachment": false, + "gpt-5.5-pro": { + "id": "gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-11-25", + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 30, + "output": 180, + "cache_read": 30 } }, - "gryphe/mythomax-l2-13b": { - "id": "gryphe/mythomax-l2-13b", - "name": "MythoMax 13B", - "attachment": false, - "reasoning": false, - "tool_call": false, + "qwen3.6-plus-free": { + "id": "qwen3.6-plus-free", + "name": "Qwen3.6 Plus Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen-free", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2024-04-25", - "last_updated": "2024-04-25", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 262144, + "output": 65536 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.06, - "output": 0.06 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "sao10k/l3-euryale-70b": { - "id": "sao10k/l3-euryale-70b", - "name": "Sao10k: Llama 3 Euryale 70B v2.1", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-06-18", - "last_updated": "2026-03-15", + "knowledge": "2025-04", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.48, - "output": 1.48 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "sao10k/l3-lunaris-8b": { - "id": "sao10k/l3-lunaris-8b", - "name": "Sao10K: Llama 3 8B Lunaris", - "attachment": false, - "reasoning": false, - "tool_call": false, + "mimo-v2-omni-free": { + "id": "mimo-v2-omni-free", + "name": "MiMo V2 Omni Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "mimo-omni-free", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-08-13", - "last_updated": "2026-03-15", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "pdf"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 262144, + "output": 64000 }, + "status": "deprecated", "cost": { - "input": 0.04, - "output": 0.05 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "sao10k/l3.3-euryale-70b": { - "id": "sao10k/l3.3-euryale-70b", - "name": "Sao10K: Llama 3.3 Euryale 70B", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-12-18", - "last_updated": "2026-03-15", + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.65, - "output": 0.75 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "sao10k/l3.1-70b-hanami-x1": { - "id": "sao10k/l3.1-70b-hanami-x1", - "name": "Sao10K: Llama 3.1 70B Hanami x1", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-01-08", - "last_updated": "2026-03-15", + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16000, - "output": 16000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 3, - "output": 3 + "input": 1.07, + "output": 8.5, + "cache_read": 0.107 } }, - "sao10k/l3.1-euryale-70b": { - "id": "sao10k/l3.1-euryale-70b", - "name": "Sao10K: Llama 3.1 Euryale 70B v2.2", - "attachment": false, - "reasoning": false, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-08-28", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0.85, - "output": 0.85 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "microsoft/wizardlm-2-8x22b": { - "id": "microsoft/wizardlm-2-8x22b", - "name": "WizardLM-2 8x22B", + "glm-5-free": { + "id": "glm-5-free", + "name": "GLM-5 Free", + "description": "Legacy model retained for compatibility with older integrations", + "family": "glm-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-04-24", - "last_updated": "2024-04-24", + "knowledge": "2025-04", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65535, - "output": 8000 + "context": 204800, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 0.62, - "output": 0.62 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "microsoft/phi-4-mini-instruct": { - "id": "microsoft/phi-4-mini-instruct", - "name": "Microsoft: Phi 4 Mini Instruct", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": true, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-10-17", - "last_updated": "2026-05-07", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 262144, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/anthropic" }, "cost": { - "input": 0.08, - "output": 0.35, - "cache_read": 0.08 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625 } }, - "microsoft/phi-4": { - "id": "microsoft/phi-4", - "name": "Microsoft: Phi 4", + "nemotron-3-super-free": { + "id": "nemotron-3-super-free", + "name": "Nemotron 3 Super Free", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2026-02", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 + "context": 204800, + "output": 128000 }, + "status": "deprecated", "cost": { - "input": 0.06, - "output": 0.14 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "poolside/laguna-m.1:free": { - "id": "poolside/laguna-m.1:free", - "name": "Poolside: Laguna M.1 (free)", - "attachment": false, + "grok-build-0.1": { + "id": "grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Grok coding model for agentic engineering, edits, and codebase workflows", + "family": "grok-build", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-05-01", + "release_date": "2026-05-20", + "last_updated": "2026-05-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 2, + "cache_read": 0.2 } }, - "poolside/laguna-xs.2:free": { - "id": "poolside/laguna-xs.2:free", - "name": "Poolside: Laguna XS.2 (free)", - "attachment": false, + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" }, "cost": { - "input": 0, - "output": 0 + "input": 1.07, + "output": 8.5, + "cache_read": 0.107 } }, - "cohere/command-r7b-12-2024": { - "id": "cohere/command-r7b-12-2024", - "name": "Cohere: Command R7B (12-2024)", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai" + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } + } + } + } + }, + "inference": { + "id": "inference", + "env": ["INFERENCE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://inference.net/v1", + "name": "Inference", + "doc": "https://inference.net/models", + "models": { + "mistral/mistral-nemo-12b-instruct": { + "id": "mistral/mistral-nemo-12b-instruct", + "name": "Mistral Nemo 12B Instruct", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-02-27", - "last_updated": "2024-02-27", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0.0375, - "output": 0.15 + "input": 0.038, + "output": 0.1 } }, - "cohere/command-a": { - "id": "cohere/command-a", - "name": "Cohere: Command A", - "attachment": false, + "google/gemma-3": { + "id": "google/gemma-3", + "name": "Google Gemma 3", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8192 + "context": 125000, + "output": 4096 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.15, + "output": 0.3 } }, - "cohere/command-r-plus-08-2024": { - "id": "cohere/command-r-plus-08-2024", - "name": "Cohere: Command R+ (08-2024)", + "osmosis/osmosis-structure-0.6b": { + "id": "osmosis/osmosis-structure-0.6b", + "name": "Osmosis Structure 0.6B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "osmosis", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 4000, + "output": 2048 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.1, + "output": 0.5 } }, - "cohere/command-r-08-2024": { - "id": "cohere/command-r-08-2024", - "name": "Cohere: Command R (08-2024)", + "qwen/qwen3-embedding-4b": { + "id": "qwen/qwen3-embedding-4b", + "name": "Qwen 3 Embedding 4B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "tool_call": false, + "temperature": false, + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 32000, + "output": 2048 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.01, + "output": 0 } }, - "prime-intellect/intellect-3": { - "id": "prime-intellect/intellect-3", - "name": "Prime Intellect: INTELLECT-3", - "attachment": false, - "reasoning": true, + "qwen/qwen-2.5-7b-vision-instruct": { + "id": "qwen/qwen-2.5-7b-vision-instruct", + "name": "Qwen 2.5 7B Vision Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-11-26", - "last_updated": "2026-02-04", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 125000, + "output": 4096 }, "cost": { "input": 0.2, - "output": 1.1 + "output": 0.2 } }, - "nvidia/llama-3.3-nemotron-super-49b-v1.5": { - "id": "nvidia/llama-3.3-nemotron-super-49b-v1.5", - "name": "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5", + "meta/llama-3.1-8b-instruct": { + "id": "meta/llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-03-16", - "last_updated": "2025-03-16", + "knowledge": "2023-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.025, + "output": 0.025 } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "NVIDIA: Nemotron 3 Super", + "meta/llama-3.2-1b-instruct": { + "id": "meta/llama-3.2-1b-instruct", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-04-11", + "knowledge": "2023-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.5, - "cache_read": 0.1 + "input": 0.01, + "output": 0.01 } }, - "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": { - "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free", - "name": "NVIDIA: Nemotron 3 Nano Omni (free)", + "meta/llama-3.2-11b-vision-instruct": { + "id": "meta/llama-3.2-11b-vision-instruct", + "name": "Llama 3.2 11B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-05-01", + "knowledge": "2023-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "audio", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 65536 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.055, + "output": 0.055 } }, - "nvidia/nemotron-3-nano-30b-a3b": { - "id": "nvidia/nemotron-3-nano-30b-a3b", - "name": "NVIDIA: Nemotron 3 Nano 30B A3B", + "meta/llama-3.2-3b-instruct": { + "id": "meta/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-12", - "last_updated": "2026-02-04", + "knowledge": "2023-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 52429 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0.05, - "output": 0.2 + "input": 0.02, + "output": 0.02 } - }, - "nvidia/nemotron-3-super-120b-a12b:free": { - "id": "nvidia/nemotron-3-super-120b-a12b:free", - "name": "NVIDIA: Nemotron 3 Super (free)", - "attachment": false, + } + } + }, + "inceptron": { + "id": "inceptron", + "env": ["INCEPTRON_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.inceptron.io/v1", + "name": "Inceptron", + "doc": "https://docs.inceptron.io", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-12", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, @@ -48634,1894 +93571,2240 @@ "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.66, + "output": 3.5, + "cache_read": 0.2, + "cache_write": 0 } }, - "nvidia/nemotron-nano-9b-v2": { - "id": "nvidia/nemotron-nano-9b-v2", - "name": "NVIDIA: Nemotron Nano 9B V2", - "attachment": false, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-08-18", - "last_updated": "2025-08-18", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.04, - "output": 0.16 + "input": 0.75, + "output": 3.5, + "cache_read": 0.2, + "cache_write": 0 } }, - "nvidia/llama-3.1-nemotron-70b-instruct": { - "id": "nvidia/llama-3.1-nemotron-70b-instruct", - "name": "NVIDIA: Llama 3.1 Nemotron 70B Instruct", - "attachment": false, - "reasoning": false, + "moonshotai/Kimi-K2.6-Fast": { + "id": "moonshotai/Kimi-K2.6-Fast", + "name": "Kimi K2.6 Fast", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-10-12", - "last_updated": "2024-10-12", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 262144 }, + "status": "alpha", "cost": { - "input": 1.2, - "output": 1.2 + "input": 1.32, + "output": 7, + "cache_read": 0.4, + "cache_write": 0 } }, - "inception/mercury-2": { - "id": "inception/mercury-2", - "name": "Inception: Mercury 2", + "zai-org/GLM-5.1-FP8": { + "id": "zai-org/GLM-5.1-FP8", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 50000 + "context": 202752, + "output": 202752 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26, + "cache_write": 0 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "OpenAI: GPT-5.1-Codex-Max", - "attachment": true, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 - } - }, - "openai/gpt-5.2-chat": { - "id": "openai/gpt-5.2-chat", - "name": "OpenAI: GPT-5.2 Chat", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.2, + "output": 4.2, + "cache_read": 0.26, + "cache_write": 0 } }, - "openai/gpt-4o-mini-search-preview": { - "id": "openai/gpt-4o-mini-search-preview", - "name": "OpenAI: GPT-4o-mini Search Preview", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-01", - "last_updated": "2025-01", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 196608, + "output": 196608 }, "cost": { "input": 0.15, - "output": 0.6 + "output": 0.9, + "cache_read": 0.05, + "cache_write": 0 } - }, - "openai/gpt-5-chat": { - "id": "openai/gpt-5-chat", - "name": "OpenAI: GPT-5 Chat", + } + } + }, + "llama": { + "id": "llama", + "env": ["LLAMA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.llama.com/compat/v1/", + "name": "Llama", + "doc": "https://llama.developer.meta.com/docs/models", + "models": { + "llama-4-scout-17b-16e-instruct-fp8": { + "id": "llama-4-scout-17b-16e-instruct-fp8", + "name": "Llama-4-Scout-17B-16E-Instruct-FP8", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-08-07", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 4096 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0, + "output": 0 } }, - "openai/gpt-4o-2024-05-13": { - "id": "openai/gpt-4o-2024-05-13", - "name": "OpenAI: GPT-4o (2024-05-13)", + "cerebras-llama-4-maverick-17b-128e-instruct": { + "id": "cerebras-llama-4-maverick-17b-128e-instruct", + "name": "Cerebras-Llama-4-Maverick-17B-128E-Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-05-13", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, "output": 4096 }, "cost": { - "input": 5, - "output": 15 + "input": 0, + "output": 0 } }, - "openai/gpt-5.3-chat": { - "id": "openai/gpt-5.3-chat", - "name": "OpenAI: GPT-5.3 Chat", + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": true, "reasoning": false, "tool_call": true, - "release_date": "2026-03-04", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 4096 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0, + "output": 0 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "OpenAI: GPT-5.2 Pro", + "groq-llama-4-maverick-17b-128e-instruct": { + "id": "groq-llama-4-maverick-17b-128e-instruct", + "name": "Groq-Llama-4-Maverick-17B-128E-Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2026-03-15", - "modalities": { - "input": ["image", "pdf", "text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 21, - "output": 168 - } - }, - "openai/gpt-4-1106-preview": { - "id": "openai/gpt-4-1106-preview", - "name": "OpenAI: GPT-4 Turbo (older v1106)", - "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2023-11-06", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, "output": 4096 }, "cost": { - "input": 10, - "output": 30 + "input": 0, + "output": 0 } }, - "openai/gpt-chat-latest": { - "id": "openai/gpt-chat-latest", - "name": "OpenAI: GPT Chat Latest", + "cerebras-llama-4-scout-17b-16e-instruct": { + "id": "cerebras-llama-4-scout-17b-16e-instruct", + "name": "Cerebras-Llama-4-Scout-17B-16E-Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "release_date": "2026-05-05", - "last_updated": "2026-05-07", - "modalities": { - "input": ["image", "pdf", "text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 - } - }, - "openai/gpt-4o-audio-preview": { - "id": "openai/gpt-4o-audio-preview", - "name": "OpenAI: GPT-4o Audio", - "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-08-15", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["audio", "text"], - "output": ["audio", "text"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 4096 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0, + "output": 0 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "OpenAI: GPT-5.5", + "llama-3.3-8b-instruct": { + "id": "llama-3.3-8b-instruct", + "name": "Llama-3.3-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2026-04-24", - "last_updated": "2026-05-01", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 + "input": 0, + "output": 0 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "OpenAI: GPT-5 Mini", + "llama-4-maverick-17b-128e-instruct-fp8": { + "id": "llama-4-maverick-17b-128e-instruct-fp8", + "name": "Llama-4-Maverick-17B-128E-Instruct-FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-08-07", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0, + "output": 0 } - }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "OpenAI: GPT-5 Nano", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-08-07", - "last_updated": "2026-03-15", + } + } + }, + "llmtr": { + "id": "llmtr", + "env": ["LLMTR_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://llmtr.com/v1", + "name": "LLMTR", + "doc": "https://llmtr.com/docs", + "models": { + "sincap": { + "id": "sincap", + "name": "Sincap", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-05-05", + "last_updated": "2026-05-05", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 0, + "output": 0 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "OpenAI: GPT-5.3-Codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "release_date": "2026-02-25", - "last_updated": "2026-03-15", + "magibu-11b-v8": { + "id": "magibu-11b-v8", + "name": "Magibu 11B v8", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-06-05", + "last_updated": "2026-06-05", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 8192, + "output": 4096 }, "cost": { - "input": 1.75, - "output": 14 + "input": 0, + "output": 0 } }, - "openai/gpt-3.5-turbo-16k": { - "id": "openai/gpt-3.5-turbo-16k", - "name": "OpenAI: GPT-3.5 Turbo 16k", + "gemma-4": { + "id": "gemma-4", + "name": "Gemma 4", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2023-08-28", - "last_updated": "2026-03-15", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16385, - "output": 4096 + "context": 32768, + "output": 8192 }, "cost": { - "input": 3, - "output": 4 + "input": 5, + "output": 10 } }, - "openai/gpt-4-turbo": { - "id": "openai/gpt-4-turbo", - "name": "OpenAI: GPT-4 Turbo", + "medgemma-4b": { + "id": "medgemma-4b", + "name": "MedGemma 4B", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2023-09-13", - "last_updated": "2024-04-09", + "release_date": "2026-04-26", + "last_updated": "2026-04-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, + "context": 8192, "output": 4096 }, "cost": { - "input": 10, - "output": 30 - } - }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "OpenAI: GPT-5.2", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2026-03-15", - "modalities": { - "input": ["image", "pdf", "text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 3, + "output": 5 } }, - "openai/o3-pro": { - "id": "openai/o3-pro", - "name": "OpenAI: o3 Pro", + "qwen3-6-35b": { + "id": "qwen3-6-35b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-04-16", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 16384, + "output": 65536 }, "cost": { - "input": 20, - "output": 80 + "input": 5, + "output": 10 } }, - "openai/o3-mini-high": { - "id": "openai/o3-mini-high", - "name": "OpenAI: o3 Mini High", - "attachment": true, + "trendyol-7b": { + "id": "trendyol-7b", + "name": "Trendyol 7B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-01-31", - "last_updated": "2026-03-15", + "tool_call": false, + "temperature": true, + "release_date": "2026-06-06", + "last_updated": "2026-06-06", "modalities": { - "input": ["pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 32768, + "output": 8192 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0, + "output": 0 } - }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "OpenAI: GPT-4o-mini", - "attachment": true, + } + } + }, + "cohere": { + "id": "cohere", + "env": ["COHERE_API_KEY"], + "npm": "@ai-sdk/cohere", + "name": "Cohere", + "doc": "https://docs.cohere.com/docs/models", + "models": { + "c4ai-aya-expanse-32b": { + "id": "c4ai-aya-expanse-32b", + "name": "Aya Expanse 32B", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2024-07-18", - "last_updated": "2026-03-15", + "release_date": "2024-10-24", + "last_updated": "2024-10-24", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 - }, - "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.075 + "output": 4000 } }, - "openai/o4-mini-deep-research": { - "id": "openai/o4-mini-deep-research", - "name": "OpenAI: o4 Mini Deep Research", - "attachment": true, - "reasoning": true, + "command-a-03-2025": { + "id": "command-a-03-2025", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-06-26", - "last_updated": "2026-03-15", + "knowledge": "2024-06-01", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 256000, + "output": 8000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 2.5, + "output": 10 } }, - "openai/gpt-5.4-mini": { - "id": "openai/gpt-5.4-mini", - "name": "OpenAI: GPT-5.4 Mini", + "c4ai-aya-vision-32b": { + "id": "c4ai-aya-vision-32b", + "name": "Aya Vision 32B", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-03-17", - "last_updated": "2026-04-11", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-03-04", + "last_updated": "2025-05-14", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "context": 16000, + "output": 4000 } }, - "openai/gpt-5.1-chat": { - "id": "openai/gpt-5.1-chat", - "name": "OpenAI: GPT-5.1 Chat", - "attachment": true, + "command-r7b-arabic-02-2025": { + "id": "command-r7b-arabic-02-2025", + "name": "Command R7B Arabic", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-13", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 - }, - "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 - } - }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "OpenAI: o4 Mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-16", - "last_updated": "2026-03-15", - "modalities": { - "input": ["image", "pdf", "text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 100000 - }, - "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.275 - } - }, - "openai/gpt-5.4-nano": { - "id": "openai/gpt-5.4-nano", - "name": "OpenAI: GPT-5.4 Nano", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-03-17", - "last_updated": "2026-04-11", - "modalities": { - "input": ["image", "pdf", "text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "output": 128000 + "output": 4000 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.0375, + "output": 0.15 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "OpenAI: GPT-5.2-Codex", + "c4ai-aya-vision-8b": { + "id": "c4ai-aya-vision-8b", + "name": "Aya Vision 8B", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-03-04", + "last_updated": "2025-05-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "context": 16000, + "output": 4000 } }, - "openai/gpt-4o-mini-2024-07-18": { - "id": "openai/gpt-4o-mini-2024-07-18", - "name": "OpenAI: GPT-4o-mini (2024-07-18)", - "attachment": true, + "command-r-08-2024": { + "id": "command-r-08-2024", + "name": "Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2024-07-18", - "last_updated": "2026-03-15", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 4000 }, "cost": { "input": 0.15, "output": 0.6 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "OpenAI: GPT-5.1-Codex-Mini", - "attachment": true, - "reasoning": true, + "command-r7b-12-2024": { + "id": "command-r7b-12-2024", + "name": "Command R7B", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-12-02", + "last_updated": "2024-12-02", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 100000 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.0375, + "output": 0.15 } }, - "openai/gpt-4o-2024-08-06": { - "id": "openai/gpt-4o-2024-08-06", - "name": "OpenAI: GPT-4o (2024-08-06)", - "attachment": true, + "command-a-vision-07-2025": { + "id": "command-a-vision-07-2025", + "name": "Command A Vision", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "release_date": "2024-08-06", - "last_updated": "2026-03-15", + "knowledge": "2024-06-01", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 8000 }, "cost": { "input": 2.5, - "output": 10, - "cache_read": 1.25 + "output": 10 } }, - "openai/gpt-5-image": { - "id": "openai/gpt-5-image", - "name": "OpenAI: GPT-5 Image", + "command-a-plus-05-2026": { + "id": "command-a-plus-05-2026", + "name": "Command A Plus", + "description": "Cohere's stronger command model for multilingual agents and enterprise workflows", + "family": "command-a", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-10-14", - "last_updated": "2026-03-15", + "knowledge": "2025-04-01", + "release_date": "2026-05-20", + "last_updated": "2026-06-09", "modalities": { - "input": ["image", "pdf", "text"], - "output": ["image", "text"] + "input": ["text", "image"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "output": 64000 }, "cost": { - "input": 10, + "input": 2.5, "output": 10 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "OpenAI: GPT-5.1", - "attachment": true, - "reasoning": true, + "command-a-translate-08-2025": { + "id": "command-a-translate-08-2025", + "name": "Command A Translate", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "family": "command-a", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-13", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 8000, + "output": 8000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 2.5, + "output": 10 } }, - "openai/o1": { - "id": "openai/o1", - "name": "OpenAI: o1", - "attachment": true, + "command-r-plus-08-2024": { + "id": "command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-12-05", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "output": 4000 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 2.5, + "output": 10 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "OpenAI: GPT-5.4 Pro", - "attachment": true, + "command-a-reasoning-08-2025": { + "id": "command-a-reasoning-08-2025", + "name": "Command A Reasoning", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1 + } + ], "tool_call": true, - "release_date": "2026-03-06", - "last_updated": "2026-03-15", + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "output": 128000 + "context": 256000, + "output": 32000 }, "cost": { - "input": 30, - "output": 180 + "input": 2.5, + "output": 10 } }, - "openai/gpt-3.5-turbo": { - "id": "openai/gpt-3.5-turbo", - "name": "OpenAI: GPT-3.5 Turbo", + "north-mini-code-1-0": { + "id": "north-mini-code-1-0", + "name": "North Mini Code", + "description": "Cohere coding model for practical software engineering and agentic edits", + "family": "north", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2023-03-01", - "last_updated": "2023-11-06", + "knowledge": "2025-09-23", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16385, - "output": 4096 + "context": 256000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.cohere.ai/compatibility/v1" }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0, + "output": 0 } }, - "openai/o3-deep-research": { - "id": "openai/o3-deep-research", - "name": "OpenAI: o3 Deep Research", - "attachment": true, - "reasoning": true, - "tool_call": true, + "c4ai-aya-expanse-8b": { + "id": "c4ai-aya-expanse-8b", + "name": "Aya Expanse 8B", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2024-06-26", - "last_updated": "2026-03-15", + "release_date": "2024-10-24", + "last_updated": "2024-10-24", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 - }, - "cost": { - "input": 10, - "output": 40, - "cache_read": 2.5 + "context": 8000, + "output": 4000 } - }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "OpenAI: o3 Mini", - "attachment": true, - "reasoning": false, + } + } + }, + "sarvam": { + "id": "sarvam", + "env": ["SARVAM_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.sarvam.ai/v1", + "name": "Sarvam AI", + "doc": "https://docs.sarvam.ai/api-reference-docs/getting-started/models", + "models": { + "sarvam-105b": { + "id": "sarvam-105b", + "name": "Sarvam-105B", + "description": "Flagship Indian-language reasoning model for enterprise multilingual applications", + "family": "sarvam", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": [null, "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2024-12-20", - "last_updated": "2026-03-15", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-18", + "last_updated": "2026-03-06", "modalities": { - "input": ["pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 - }, - "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "context": 131072, + "output": 131072 } }, - "openai/gpt-4-turbo-preview": { - "id": "openai/gpt-4-turbo-preview", - "name": "OpenAI: GPT-4 Turbo Preview", + "sarvam-30b": { + "id": "sarvam-30b", + "name": "Sarvam-30B", + "description": "Efficient Indian-language reasoning model for chat, coding, and multilingual work", + "family": "sarvam", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": [null, "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-01-25", - "last_updated": "2026-03-15", + "release_date": "2026-02-18", + "last_updated": "2026-03-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 - }, - "cost": { - "input": 10, - "output": 30 + "context": 65536, + "output": 65536 } - }, - "openai/o1-pro": { - "id": "openai/o1-pro", - "name": "OpenAI: o1-pro", - "attachment": true, + } + } + }, + "stepfun": { + "id": "stepfun", + "env": ["STEPFUN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.stepfun.com/v1", + "name": "StepFun", + "doc": "https://platform.stepfun.com/docs/zh/overview/concept", + "models": { + "step-1-32k": { + "id": "step-1-32k", + "name": "Step 1 (32K)", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": false, "reasoning": true, - "tool_call": false, - "temperature": false, - "release_date": "2025-03-19", - "last_updated": "2026-03-15", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-01-01", + "last_updated": "2026-02-13", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 150, - "output": 600 + "input": 2.05, + "output": 9.59, + "cache_read": 0.41 } }, - "openai/gpt-5.4-image-2": { - "id": "openai/gpt-5.4-image-2", - "name": "OpenAI: GPT-5.4 Image 2", + "step-3.7-flash": { + "id": "step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": false, - "release_date": "2026-04-21", - "last_updated": "2026-05-01", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-06-29", "modalities": { - "input": ["image", "text", "pdf"], - "output": ["image", "text"] + "input": ["text", "image"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 272000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 8, - "output": 15, - "cache_read": 2 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "openai/gpt-4": { - "id": "openai/gpt-4", - "name": "OpenAI: GPT-4", + "step-3.5-flash-2603": { + "id": "step-3.5-flash-2603", + "name": "Step 3.5 Flash 2603", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2023-03-14", - "last_updated": "2024-04-09", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8191, - "output": 4096 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 30, - "output": 60 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "openai/gpt-4-0314": { - "id": "openai/gpt-4-0314", - "name": "OpenAI: GPT-4 (older v0314)", + "stepaudio-2.5-tts": { + "id": "stepaudio-2.5-tts", + "name": "StepAudio 2.5 TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "step", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2023-05-28", - "last_updated": "2026-03-15", + "tool_call": false, + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-07-02", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 4096 - }, - "cost": { - "input": 30, - "output": 60 + "context": 0, + "output": 0 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "OpenAI: GPT-5 Codex", - "attachment": true, - "reasoning": true, - "tool_call": true, + "stepaudio-2.5-asr": { + "id": "stepaudio-2.5-asr", + "name": "StepAudio 2.5 ASR", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "step", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": false, - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "release_date": "2026-04-24", + "last_updated": "2026-07-02", "modalities": { - "input": ["text", "image"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "context": 0, + "output": 0 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "OpenAI: GPT-5.4", - "attachment": true, + "step-3.5-flash": { + "id": "step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2026-03-06", - "last_updated": "2026-03-15", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01-29", + "last_updated": "2026-06-15", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 2.5, - "output": 15 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "openai/gpt-audio": { - "id": "openai/gpt-audio", - "name": "OpenAI: GPT Audio", + "step-tts-2": { + "id": "step-tts-2", + "name": "Step TTS 2", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "step", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2026-01-20", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-03-01", + "last_updated": "2026-07-02", "modalities": { - "input": ["audio", "text"], - "output": ["audio", "text"] + "input": ["text"], + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 - }, - "cost": { - "input": 2.5, - "output": 10 + "context": 0, + "output": 0 } }, - "openai/gpt-4o-search-preview": { - "id": "openai/gpt-4o-search-preview", - "name": "OpenAI: GPT-4o Search Preview", + "step-2-16k": { + "id": "step-2-16k", + "name": "Step 2 (16K)", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2025-03-13", - "last_updated": "2026-03-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-01-01", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 10 + "input": 5.21, + "output": 16.44, + "cache_read": 1.04 } - }, - "openai/gpt-4.1-nano": { - "id": "openai/gpt-4.1-nano", - "name": "OpenAI: GPT-4.1 Nano", + } + } + }, + "hpc-ai": { + "id": "hpc-ai", + "env": ["HPC_AI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.hpc-ai.com/inference/v1", + "name": "HPC-AI", + "doc": "https://www.hpc-ai.com/doc/docs/quickstart/", + "models": { + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "openai/o4-mini-high": { - "id": "openai/o4-mini-high", - "name": "OpenAI: o4 Mini High", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "release_date": "2025-04-17", - "last_updated": "2026-03-15", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 1.1, - "output": 4.4 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "openai/o3": { - "id": "openai/o3", - "name": "OpenAI: o3", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-04-16", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": true, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "OpenAI: gpt-oss-20b", - "attachment": false, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.03, - "output": 0.14 + "input": 5, + "output": 25, + "cache_read": 0.5 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "OpenAI: GPT-5 Pro", - "attachment": true, + "zai-org/glm-5.1": { + "id": "zai-org/glm-5.1", + "name": "GLM 5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-06", - "last_updated": "2026-03-15", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-08", + "last_updated": "2026-06-01", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 202000, + "output": 202000 }, "cost": { - "input": 15, - "output": 120 + "input": 0.615, + "output": 2.46, + "cache_read": 0.133 } }, - "openai/gpt-audio-mini": { - "id": "openai/gpt-audio-mini", - "name": "OpenAI: GPT Audio Mini", + "zai-org/glm-5.2": { + "id": "zai-org/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-01-20", - "last_updated": "2026-03-15", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["audio", "text"], - "output": ["audio", "text"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 2.4 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "OpenAI: GPT-4o", - "attachment": true, - "reasoning": false, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-05-13", - "last_updated": "2026-03-15", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 128000 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "openai/gpt-3.5-turbo-0613": { - "id": "openai/gpt-3.5-turbo-0613", - "name": "OpenAI: GPT-3.5 Turbo (older v0613)", + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2023-06-13", - "last_updated": "2023-06-13", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 4095, - "output": 4096 + "context": 1002000, + "output": 128000 }, "cost": { - "input": 1, - "output": 2 + "input": 1.74, + "output": 3.48, + "cache_read": 0.145 } }, - "openai/gpt-5-image-mini": { - "id": "openai/gpt-5-image-mini", - "name": "OpenAI: GPT-5 Image Mini", - "attachment": true, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-10-16", - "last_updated": "2026-03-15", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["image", "pdf", "text"], - "output": ["image", "text"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 196000, + "output": 195000 }, "cost": { - "input": 2.5, - "output": 2 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } - }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "OpenAI: GPT-5", - "attachment": true, + } + } + }, + "minimax-cn": { + "id": "minimax-cn", + "env": ["MINIMAX_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.minimaxi.com/anthropic/v1", + "name": "MiniMax (minimaxi.com)", + "doc": "https://platform.minimaxi.com/docs/guides/quickstart", + "models": { + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-08-07", - "last_updated": "2026-03-15", + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 1.2 } }, - "openai/gpt-oss-safeguard-20b": { - "id": "openai/gpt-oss-safeguard-20b", - "name": "OpenAI: gpt-oss-safeguard-20b", + "MiniMax-M2.5-highspeed": { + "id": "MiniMax-M2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-10-29", - "last_updated": "2025-10-29", + "release_date": "2026-02-13", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.075, - "output": 0.3, - "cache_read": 0.037 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "OpenAI: gpt-oss-120b", + "MiniMax-M2.7-highspeed": { + "id": "MiniMax-M2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 26215 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.039, - "output": 0.19 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "openai/gpt-5.5-pro": { - "id": "openai/gpt-5.5-pro", - "name": "OpenAI: GPT-5.5 Pro", - "attachment": true, + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2026-04-24", - "last_updated": "2026-05-01", + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, + "context": 196608, "output": 128000 }, "cost": { - "input": 30, - "output": 180 + "input": 0.3, + "output": 1.2 } }, - "openai/gpt-3.5-turbo-instruct": { - "id": "openai/gpt-3.5-turbo-instruct", - "name": "OpenAI: GPT-3.5 Turbo Instruct", + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2023-03-01", - "last_updated": "2023-09-21", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 4095, - "output": 4096 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.5, - "output": 2 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "OpenAI: GPT-4.1", + "MiniMax-M3": { + "id": "MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2026-03-15", + "release_date": "2026-06-01", + "last_updated": "2026-06-25", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "tiers": [ + { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 + } } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "OpenAI: GPT-4.1 Mini", - "attachment": true, - "reasoning": false, + "MiniMax-M2.7": { + "id": "MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2026-03-15", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } - }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "OpenAI: GPT-5.1-Codex", + } + } + }, + "unorouter": { + "id": "unorouter", + "env": ["UNOROUTER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.unorouter.com/v1", + "name": "UnoRouter", + "doc": "https://unorouter.com/models", + "models": { + "gpt-5.5:free": { + "id": "gpt-5.5:free", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0, + "output": 0 } }, - "openai/gpt-4o-2024-11-20": { - "id": "openai/gpt-4o-2024-11-20", - "name": "OpenAI: GPT-4o (2024-11-20)", - "attachment": true, - "reasoning": false, + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-11-20", - "last_updated": "2026-03-15", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.0625, + "output": 0.125 } }, - "amazon/nova-lite-v1": { - "id": "amazon/nova-lite-v1", - "name": "Amazon: Nova Lite 1.0", + "gpt-5.4:free": { + "id": "gpt-5.4:free", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2024-12-06", - "last_updated": "2026-03-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 5120 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.06, - "output": 0.24 + "input": 0, + "output": 0 } }, - "amazon/nova-pro-v1": { - "id": "amazon/nova-pro-v1", - "name": "Amazon: Nova Pro 1.0", - "attachment": true, - "reasoning": false, + "glm-4.5-flash:free": { + "id": "glm-4.5-flash:free", + "name": "GLM-4.5-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 5120 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.8, - "output": 3.2 + "input": 0, + "output": 0 } }, - "amazon/nova-premier-v1": { - "id": "amazon/nova-premier-v1", - "name": "Amazon: Nova Premier 1.0", - "attachment": true, - "reasoning": false, + "minimax-m2.7:free": { + "id": "minimax-m2.7:free", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-11-01", - "last_updated": "2026-03-15", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 12.5 + "input": 0, + "output": 0 } }, - "amazon/nova-2-lite-v1": { - "id": "amazon/nova-2-lite-v1", - "name": "Amazon: Nova 2 Lite", + "step-3.7-flash:free": { + "id": "step-3.7-flash:free", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2026-03-15", + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 65535 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 0, + "output": 0 } }, - "amazon/nova-micro-v1": { - "id": "amazon/nova-micro-v1", - "name": "Amazon: Nova Micro 1.0", - "attachment": false, - "reasoning": false, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2024-12-06", - "last_updated": "2026-03-15", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 5120 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.035, - "output": 0.14 + "input": 1.2, + "output": 6 } }, - "z-ai/glm-5v-turbo": { - "id": "z-ai/glm-5v-turbo", - "name": "Z.ai: GLM 5V Turbo", + "qwen3.5-397b-a17b:free": { + "id": "qwen3.5-397b-a17b:free", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-11", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24 + "input": 0, + "output": 0 } }, - "z-ai/glm-4.7": { - "id": "z-ai/glm-4.7", - "name": "Z.ai: GLM 4.7", + "nemotron-3-ultra-550b-a55b:free": { + "id": "nemotron-3-ultra-550b-a55b:free", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-12-22", - "last_updated": "2026-03-15", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 65535 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.38, - "output": 1.98, - "cache_read": 0.2 + "input": 0, + "output": 0 } }, - "z-ai/glm-5": { - "id": "z-ai/glm-5", - "name": "Z.ai: GLM 5", - "attachment": false, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-03-15", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.72, - "output": 2.3 + "input": 0.1857, + "output": 1.1142 } }, - "z-ai/glm-4-32b": { - "id": "z-ai/glm-4-32b", - "name": "Z.ai: GLM 4 32B ", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-07-25", - "last_updated": "2026-03-15", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.8999, + "output": 1.7999 } }, - "z-ai/glm-5.1": { - "id": "z-ai/glm-5.1", - "name": "Z.ai: GLM 5.1", + "deepseek-v4-flash:free": { + "id": "deepseek-v4-flash:free", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 1.26, - "output": 3.96 + "input": 0, + "output": 0 } }, - "z-ai/glm-4.5": { - "id": "z-ai/glm-4.5", - "name": "Z.ai: GLM 4.5", - "attachment": false, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2026-03-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.175 + "input": 1.44, + "output": 7.2 } }, - "z-ai/glm-4.5-air": { - "id": "z-ai/glm-4.5-air", - "name": "Z.ai: GLM 4.5 Air", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.13, - "output": 0.85, - "cache_read": 0.025 + "input": 1.6001, + "output": 5.0288 } }, - "z-ai/glm-5-turbo": { - "id": "z-ai/glm-5-turbo", - "name": "Z.ai: GLM 5 Turbo", - "attachment": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-04-11", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24 + "input": 0.425, + "output": 2.125 } }, - "z-ai/glm-4.5v": { - "id": "z-ai/glm-4.5v", - "name": "Z.ai: GLM 4.5V", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 1.8, - "cache_read": 0.11 + "input": 1.05, + "output": 8.4 } }, - "z-ai/glm-4.6": { - "id": "z-ai/glm-4.6", - "name": "Z.ai: GLM 4.6", + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2026-03-15", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] @@ -50529,3991 +95812,4507 @@ "open_weights": true, "limit": { "context": 204800, - "output": 204800 + "output": 131072 }, "cost": { - "input": 0.39, - "output": 1.9, - "cache_read": 0.175 + "input": 0.819, + "output": 3.276 } }, - "z-ai/glm-4.6v": { - "id": "z-ai/glm-4.6v", - "name": "Z.ai: GLM 4.6V", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-09-30", - "last_updated": "2026-01-10", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 1.8, + "output": 10.8 } }, - "z-ai/glm-4.7-flash": { - "id": "z-ai/glm-4.7-flash", - "name": "Z.ai: GLM 4.7 Flash", - "attachment": false, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 40551 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.06, - "output": 0.4, - "cache_read": 0.01 + "input": 1.2675, + "output": 5.3368 } }, - "baidu/ernie-4.5-vl-424b-a47b": { - "id": "baidu/ernie-4.5-vl-424b-a47b", - "name": "Baidu: ERNIE 4.5 VL 424B A47B ", - "attachment": true, + "glm-5.2:free": { + "id": "glm-5.2:free", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2026-01", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 123000, - "output": 16000 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.42, - "output": 1.25 + "input": 0, + "output": 0 } }, - "baidu/qianfan-ocr-fast:free": { - "id": "baidu/qianfan-ocr-fast:free", - "name": "Baidu: Qianfan-OCR-Fast (free)", + "gemma-4-31b-it:free": { + "id": "gemma-4-31b-it:free", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-05-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "output": 28672 + "context": 262144, + "output": 32768 }, "cost": { "input": 0, "output": 0 } }, - "baidu/cobuddy:free": { - "id": "baidu/cobuddy:free", - "name": "Baidu: CoBuddy (free)", + "deepseek-v4-pro:free": { + "id": "deepseek-v4-pro:free", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2026-05-06", - "last_updated": "2026-05-07", + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 1000000, + "output": 384000 }, "cost": { "input": 0, "output": 0 } }, - "baidu/ernie-4.5-vl-28b-a3b": { - "id": "baidu/ernie-4.5-vl-28b-a3b", - "name": "Baidu: ERNIE 4.5 VL 28B A3B", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 30000, - "output": 8000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.56 + "input": 0.1875, + "output": 1.125 } - }, - "baidu/ernie-4.5-21b-a3b": { - "id": "baidu/ernie-4.5-21b-a3b", - "name": "Baidu: ERNIE 4.5 21B A3B", + } + } + }, + "alibaba-coding-plan": { + "id": "alibaba-coding-plan", + "env": ["ALIBABA_CODING_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://coding-intl.dashscope.aliyuncs.com/v1", + "name": "Alibaba Coding Plan", + "doc": "https://www.alibabacloud.com/help/en/model-studio/coding-plan", + "models": { + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-06-30", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 120000, - "output": 8000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "baidu/ernie-4.5-300b-a47b": { - "id": "baidu/ernie-4.5-300b-a47b", - "name": "Baidu: ERNIE 4.5 300B A47B ", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2026-01", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 123000, - "output": 12000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "baidu/ernie-4.5-21b-a3b-thinking": { - "id": "baidu/ernie-4.5-21b-a3b-thinking", - "name": "Baidu: ERNIE 4.5 21B A3B Thinking", + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "relace/relace-apply-3": { - "id": "relace/relace-apply-3", - "name": "Relace: Relace Apply 3", + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2025-09-26", - "last_updated": "2026-03-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.85, - "output": 1.25 + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 } }, - "relace/relace-search": { - "id": "relace/relace-search", - "name": "Relace: Relace Search", - "attachment": false, - "reasoning": false, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-12-09", - "last_updated": "2026-03-15", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 1, - "output": 3 + "input": 0.1875, + "output": 1.125, + "cache_write": 0.234375 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "MiniMax: MiniMax M2.7", - "family": "minimax-m2.7", + "qwen3-max-2026-01-23": { + "id": "qwen3-max-2026-01-23", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-04", + "release_date": "2026-01-23", + "last_updated": "2026-01-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "MiniMax: MiniMax M2", + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "release_date": "2025-10-23", - "last_updated": "2026-03-15", + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 196608 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.255, - "output": 1, - "cache_read": 0.03 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "minimax/minimax-01": { - "id": "minimax/minimax-01", - "name": "MiniMax: MiniMax-01", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-01-15", - "last_updated": "2025-01-15", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 1000192, - "output": 1000192 - }, - "cost": { - "input": 0.2, - "output": 1.1 - } - }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "MiniMax: MiniMax M2.1", - "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2025-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 39322 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.27, - "output": 0.95, - "cache_read": 0.03 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "minimax/minimax-m1": { - "id": "minimax/minimax-m1", - "name": "MiniMax: MiniMax M1", + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 40000 + "context": 196608, + "input": 196601, + "output": 24576 }, "cost": { - "input": 0.4, - "output": 2.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "minimax/minimax-m2-her": { - "id": "minimax/minimax-m2-her", - "name": "MiniMax: MiniMax M2-her", + "qwen3-coder-next": { + "id": "qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-01-23", - "last_updated": "2026-03-15", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 2048 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax: MiniMax M2.5", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-03-15", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 196608, - "output": 196608 - }, - "cost": { - "input": 0.25, - "output": 1.2, - "cache_read": 0.029 - } - }, - "~openai/gpt-latest": { - "id": "~openai/gpt-latest", - "name": "OpenAI: GPT Latest", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, "open_weights": false, "limit": { - "context": 1050000, - "output": 128000 + "context": 202752, + "output": 16384 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "~openai/gpt-mini-latest": { - "id": "~openai/gpt-mini-latest", - "name": "OpenAI: GPT Mini Latest", - "attachment": true, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "qwen/qwen3-235b-a22b": { - "id": "qwen/qwen3-235b-a22b", - "name": "Qwen: Qwen3 235B A22B", + } + } + }, + "longcat": { + "id": "longcat", + "env": ["LONGCAT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.longcat.chat/openai", + "name": "LongCat", + "doc": "https://longcat.chat/platform/docs/", + "models": { + "LongCat-2.0": { + "id": "LongCat-2.0", + "name": "LongCat-2.0", + "description": "Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window", + "family": "longcat", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2026-03-15", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.455, - "output": 1.82, - "cache_read": 0.15 + "input": 0.75, + "output": 2.95, + "cache_read": 0.015 } - }, - "qwen/qwen3.5-122b-a10b": { - "id": "qwen/qwen3.5-122b-a10b", - "name": "Qwen: Qwen3.5-122B-A10B", + } + } + }, + "poe": { + "id": "poe", + "env": ["POE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.poe.com/v1", + "name": "Poe", + "doc": "https://creator.poe.com/docs/external-applications/openai-compatible-api", + "models": { + "trytako/tako": { + "id": "trytako/tako", + "name": "Tako", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "tako", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { - "input": ["image", "text", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.26, - "output": 2.08 + "context": 2048, + "output": 0 } }, - "qwen/qwen3-coder-plus": { - "id": "qwen/qwen3-coder-plus", - "name": "Qwen: Qwen3 Coder Plus", - "attachment": false, - "reasoning": false, + "xai/grok-code-fast-1": { + "id": "xai/grok-code-fast-1", + "name": "Grok Code Fast 1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-07-01", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-08-22", + "last_updated": "2025-08-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.65, - "output": 3.25, - "cache_read": 0.2 + "input": 0.2, + "output": 1.5, + "cache_read": 0.02 } }, - "qwen/qwen3.6-27b": { - "id": "qwen/qwen3.6-27b", - "name": "Qwen: Qwen3.6 27B", + "xai/grok-4.1-fast-reasoning": { + "id": "xai/grok-4.1-fast-reasoning", + "name": "Grok-4.1-Fast-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": false, + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 65536 - }, - "cost": { - "input": 0.325, - "output": 3.25 + "context": 2000000, + "output": 30000 } }, - "qwen/qwen3.5-27b": { - "id": "qwen/qwen3.5-27b", - "name": "Qwen: Qwen3.5-27B", + "xai/grok-3-mini": { + "id": "xai/grok-3-mini", + "name": "Grok 3 Mini", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { - "input": ["image", "text", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.195, - "output": 1.56 + "input": 0.3, + "output": 0.5, + "cache_read": 0.075 } }, - "qwen/qwen3-235b-a22b-2507": { - "id": "qwen/qwen3-235b-a22b-2507", - "name": "Qwen: Qwen3 235B A22B Instruct 2507", - "attachment": false, - "reasoning": true, + "xai/grok-4.1-fast-non-reasoning": { + "id": "xai/grok-4.1-fast-non-reasoning", + "name": "Grok-4.1-Fast-Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-04", - "last_updated": "2026-01", + "temperature": false, + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 52429 - }, - "cost": { - "input": 0.071, - "output": 0.1 + "context": 2000000, + "output": 30000 } }, - "qwen/qwen3-8b": { - "id": "qwen/qwen3-8b", - "name": "Qwen: Qwen3 8B", - "attachment": false, - "reasoning": true, + "xai/grok-3": { + "id": "xai/grok-3", + "name": "Grok 3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-04", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-04-11", + "last_updated": "2025-04-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, + "context": 131072, "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.05 + "input": 3, + "output": 15, + "cache_read": 0.75 } }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen: Qwen3.5 397B A17B", + "xai/grok-4-fast-reasoning": { + "id": "xai/grok-4-fast-reasoning", + "name": "Grok-4-Fast-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-09-16", + "last_updated": "2025-09-16", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 2000000, + "output": 128000 }, "cost": { - "input": 0.39, - "output": 2.34 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "qwen/qwen-vl-plus": { - "id": "qwen/qwen-vl-plus", - "name": "Qwen: Qwen VL Plus", + "xai/grok-4": { + "id": "xai/grok-4", + "name": "Grok-4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-01-25", - "last_updated": "2026-03-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "release_date": "2025-07-10", + "last_updated": "2025-07-10", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.1365, - "output": 0.4095, - "cache_read": 0.042 + "input": 3, + "output": 15, + "cache_read": 0.75 } }, - "qwen/qwen3-32b": { - "id": "qwen/qwen3-32b", - "name": "Qwen: Qwen3 32B", - "attachment": false, - "reasoning": true, + "xai/grok-4-fast-non-reasoning": { + "id": "xai/grok-4-fast-non-reasoning", + "name": "Grok-4-Fast-Non-Reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2026-02-04", + "temperature": false, + "release_date": "2025-09-16", + "last_updated": "2025-09-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 2000000, + "output": 128000 }, "cost": { - "input": 0.08, - "output": 0.24, - "cache_read": 0.04 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "qwen/qwen2.5-vl-72b-instruct": { - "id": "qwen/qwen2.5-vl-72b-instruct", - "name": "Qwen: Qwen2.5 VL 72B Instruct", + "xai/grok-4.20-multi-agent": { + "id": "xai/grok-4.20-multi-agent", + "name": "Grok-4.20-Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-02-01", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2026-03-13", + "last_updated": "2026-03-13", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 0 }, "cost": { - "input": 0.8, - "output": 0.8, - "cache_read": 0.075 + "input": 2, + "output": 6, + "cache_read": 0.2 } }, - "qwen/qwen-max": { - "id": "qwen/qwen-max", - "name": "Qwen: Qwen-Max ", - "attachment": false, + "topazlabs-co/topazlabs": { + "id": "topazlabs-co/topazlabs", + "name": "TopazLabs", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "topazlabs", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-04-03", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 8192 - }, - "cost": { - "input": 1.04, - "output": 4.16, - "cache_read": 0.32 + "context": 204, + "output": 0 } }, - "qwen/qwen-plus": { - "id": "qwen/qwen-plus", - "name": "Qwen: Qwen-Plus", - "attachment": false, + "fireworks-ai/kimi-k2.5-fw": { + "id": "fireworks-ai/kimi-k2.5-fw", + "name": "Kimi-K2.5-FW", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-01-25", - "last_updated": "2025-09-11", + "temperature": false, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 262144, + "input": 245760, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 1.2, - "cache_read": 0.08 + "input": 0, + "output": 0 } }, - "qwen/qwen3.6-35b-a3b": { - "id": "qwen/qwen3.6-35b-a3b", - "name": "Qwen: Qwen3.6 35B A3B", + "google/veo-3.1-fast": { + "id": "google/veo-3.1-fast", + "name": "Veo-3.1-Fast", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": true, - "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.1612, - "output": 0.96525, - "cache_read": 0.1612 + "context": 480, + "output": 0 } }, - "qwen/qwen3-vl-235b-a22b-thinking": { - "id": "qwen/qwen3-vl-235b-a22b-thinking", - "name": "Qwen: Qwen3 VL 235B A22B Thinking", + "google/imagen-3": { + "id": "google/imagen-3", + "name": "Imagen-3", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-10-15", + "last_updated": "2024-10-15", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - }, - "cost": { - "input": 0.26, - "output": 2.6 + "context": 480, + "output": 0 } }, - "qwen/qwen3-vl-30b-a3b-thinking": { - "id": "qwen/qwen3-vl-30b-a3b-thinking", - "name": "Qwen: Qwen3 VL 30B A3B Thinking", + "google/nano-banana-pro": { + "id": "google/nano-banana-pro", + "name": "Nano-Banana-Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "nano-banana", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-11-19", + "last_updated": "2025-11-19", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 65536, + "output": 0 }, "cost": { - "input": 0.13, - "output": 1.56 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "qwen/qwen3-vl-8b-instruct": { - "id": "qwen/qwen3-vl-8b-instruct", - "name": "Qwen: Qwen3 VL 8B Instruct", + "google/lyria": { + "id": "google/lyria", + "name": "Lyria", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "lyria", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "temperature": false, + "release_date": "2025-06-04", + "last_updated": "2025-06-04", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text"], + "output": ["audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - }, - "cost": { - "input": 0.08, - "output": 0.5 + "context": 0, + "output": 0 } }, - "qwen/qwen3.5-flash-02-23": { - "id": "qwen/qwen3.5-flash-02-23", - "name": "Qwen: Qwen3.5-Flash", + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini-3.1-Flash-Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-02-18", + "last_updated": "2026-02-18", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, + "context": 1048576, "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.25, + "output": 1.5 } }, - "qwen/qwen3.6-plus": { - "id": "qwen/qwen3.6-plus", - "name": "Qwen: Qwen3.6 Plus", + "google/nano-banana": { + "id": "google/nano-banana", + "name": "Nano-Banana", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "nano-banana", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-08-26", - "last_updated": "2026-04-11", + "temperature": false, + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 65536, + "output": 0 }, "cost": { - "input": 0.325, - "output": 1.95, - "cache_read": 0.0325, - "cache_write": 0.40625 + "input": 0.21, + "output": 1.8, + "cache_read": 0.021 } }, - "qwen/qwen3-max": { - "id": "qwen/qwen3-max", - "name": "Qwen: Qwen3 Max", - "attachment": false, - "reasoning": false, + "google/gemini-deep-research": { + "id": "google/gemini-deep-research", + "name": "gemini-deep-research", + "description": "Legacy model retained for compatibility with older integrations", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-09-05", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1048576, + "output": 0 }, + "status": "deprecated", "cost": { - "input": 1.2, - "output": 6, - "cache_read": 0.24 + "input": 1.6, + "output": 9.6 } }, - "qwen/qwen-plus-2025-07-28": { - "id": "qwen/qwen-plus-2025-07-28", - "name": "Qwen: Qwen Plus 0728", - "attachment": false, + "google/veo-3": { + "id": "google/veo-3", + "name": "Veo-3", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-09", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-05-21", + "last_updated": "2025-05-21", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 - }, - "cost": { - "input": 0.26, - "output": 0.78 + "context": 480, + "output": 0 } }, - "qwen/qwen3-30b-a3b-instruct-2507": { - "id": "qwen/qwen3-30b-a3b-instruct-2507", - "name": "Qwen: Qwen3 30B A3B Instruct 2507", - "attachment": false, - "reasoning": false, + "google/gemini-3-flash": { + "id": "google/gemini-3-flash", + "name": "Gemini-3-Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-29", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-07", + "last_updated": "2025-10-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.09, - "output": 0.3, + "input": 0.4, + "output": 2.4, "cache_read": 0.04 } }, - "qwen/qwen3-vl-32b-instruct": { - "id": "qwen/qwen3-vl-32b-instruct", - "name": "Qwen: Qwen3 VL 32B Instruct", + "google/imagen-3-fast": { + "id": "google/imagen-3-fast", + "name": "Imagen-3-Fast", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-10-21", - "last_updated": "2025-11-25", + "temperature": false, + "release_date": "2024-10-17", + "last_updated": "2024-10-17", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - }, - "cost": { - "input": 0.104, - "output": 0.416 + "context": 480, + "output": 0 } }, - "qwen/qwen3-235b-a22b-thinking-2507": { - "id": "qwen/qwen3-235b-a22b-thinking-2507", - "name": "Qwen: Qwen3 235B A22B Thinking 2507", - "attachment": false, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini-2.5-Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 32768 + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-25", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-02-05", + "last_updated": "2025-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1065535, + "output": 65535 }, "cost": { - "input": 0.11, - "output": 0.6 + "input": 0.87, + "output": 7, + "cache_read": 0.087 } }, - "qwen/qwen3-next-80b-a3b-thinking": { - "id": "qwen/qwen3-next-80b-a3b-thinking", - "name": "Qwen: Qwen3 Next 80B A3B Thinking", - "attachment": false, - "reasoning": true, + "google/veo-3.1": { + "id": "google/veo-3.1", + "name": "Veo-3.1", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 - }, - "cost": { - "input": 0.0975, - "output": 0.78 + "context": 480, + "output": 0 } }, - "qwen/qwen3-30b-a3b-thinking-2507": { - "id": "qwen/qwen3-30b-a3b-thinking-2507", - "name": "Qwen: Qwen3 30B A3B Thinking 2507", - "attachment": false, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini-2.5-Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "temperature": false, + "release_date": "2025-04-26", + "last_updated": "2025-04-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 6554 + "context": 1065535, + "output": 65535 }, "cost": { - "input": 0.051, - "output": 0.34 + "input": 0.21, + "output": 1.8, + "cache_read": 0.021 } }, - "qwen/qwen-2.5-7b-instruct": { - "id": "qwen/qwen-2.5-7b-instruct", - "name": "Qwen: Qwen2.5 7B Instruct", - "attachment": false, + "google/imagen-4-fast": { + "id": "google/imagen-4-fast", + "name": "Imagen-4-Fast", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-09", - "last_updated": "2025-04-16", + "temperature": false, + "release_date": "2025-06-25", + "last_updated": "2025-06-25", "modalities": { "input": ["text"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 480, + "output": 0 + } + }, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini-3.5-Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 6554 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.04, - "output": 0.1 + "input": 1.5152, + "output": 9.0909, + "cache_read": 0.1515 } }, - "qwen/qwen-vl-max": { - "id": "qwen/qwen-vl-max", - "name": "Qwen: Qwen VL Max", + "google/gemma-4-31b": { + "id": "google/gemma-4-31b", + "name": "Gemma-4-31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-04-08", - "last_updated": "2025-08-13", + "temperature": false, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "output": 8192 }, "cost": { - "input": 0.8, - "output": 3.2 + "input": 0, + "output": 0 } }, - "qwen/qwen3-coder-flash": { - "id": "qwen/qwen3-coder-flash", - "name": "Qwen: Qwen3 Coder Flash", - "attachment": false, + "google/veo-3-fast": { + "id": "google/veo-3-fast", + "name": "Veo-3-Fast", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-13", + "last_updated": "2025-10-13", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 - }, - "cost": { - "input": 0.195, - "output": 0.975, - "cache_read": 0.06 + "context": 480, + "output": 0 } }, - "qwen/qwen3-30b-a3b": { - "id": "qwen/qwen3-30b-a3b", - "name": "Qwen: Qwen3 30B A3B", - "attachment": false, + "google/gemini-3-pro": { + "id": "google/gemini-3-pro", + "name": "Gemini-3-Pro", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-04", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-22", + "last_updated": "2025-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 1048576, + "output": 65536 }, + "status": "deprecated", "cost": { - "input": 0.08, - "output": 0.28, - "cache_read": 0.03 + "input": 1.6, + "output": 9.6, + "cache_read": 0.16 } }, - "qwen/qwen3-next-80b-a3b-instruct": { - "id": "qwen/qwen3-next-80b-a3b-instruct", - "name": "Qwen: Qwen3 Next 80B A3B Instruct", - "attachment": false, + "google/gemini-2.0-flash": { + "id": "google/gemini-2.0-flash", + "name": "Gemini-2.0-Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 52429 + "context": 990000, + "output": 8192 }, "cost": { - "input": 0.09, - "output": 1.1 + "input": 0.1, + "output": 0.42 } }, - "qwen/qwen3.5-plus-20260420": { - "id": "qwen/qwen3.5-plus-20260420", - "name": "Qwen: Qwen3.5 Plus 2026-04-20", + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini-2.5-Flash-Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": false, + "release_date": "2025-06-19", + "last_updated": "2025-06-19", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 1024000, + "output": 64000 }, "cost": { - "input": 0.4, - "output": 2.4 + "input": 0.07, + "output": 0.28 } }, - "qwen/qwen3-coder-next": { - "id": "qwen/qwen3-coder-next", - "name": "Qwen: Qwen3 Coder Next", - "attachment": false, + "google/imagen-4": { + "id": "google/imagen-4", + "name": "Imagen-4", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-02-02", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - }, - "cost": { - "input": 0.12, - "output": 0.75, - "cache_read": 0.035 + "context": 480, + "output": 0 } }, - "qwen/qwen-2.5-coder-32b-instruct": { - "id": "qwen/qwen-2.5-coder-32b-instruct", - "name": "Qwen2.5 Coder 32B Instruct", - "attachment": false, + "google/imagen-4-ultra": { + "id": "google/imagen-4-ultra", + "name": "Imagen-4-Ultra", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-11-11", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2025-05-24", + "last_updated": "2025-05-24", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 - }, - "cost": { - "input": 0.2, - "output": 0.2, - "cache_read": 0.015 + "context": 480, + "output": 0 } }, - "qwen/qwen3-vl-30b-a3b-instruct": { - "id": "qwen/qwen3-vl-30b-a3b-instruct", - "name": "Qwen: Qwen3 VL 30B A3B Instruct", + "google/gemini-3.1-pro": { + "id": "google/gemini-3.1-pro", + "name": "Gemini-3.1-Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-10-05", - "last_updated": "2025-11-25", + "temperature": false, + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "qwen/qwen3-coder-30b-a3b-instruct": { - "id": "qwen/qwen3-coder-30b-a3b-instruct", - "name": "Qwen: Qwen3 Coder 30B A3B Instruct", - "attachment": false, + "google/veo-2": { + "id": "google/veo-2", + "name": "Veo-2", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-07-31", - "last_updated": "2025-07-31", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 160000, - "output": 32768 - }, - "cost": { - "input": 0.07, - "output": 0.27 - } - }, - "qwen/qwen3-max-thinking": { - "id": "qwen/qwen3-max-thinking", - "name": "Qwen: Qwen3 Max Thinking", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-01-23", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-12-02", + "last_updated": "2024-12-02", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["video"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 - }, - "cost": { - "input": 0.78, - "output": 3.9 + "context": 480, + "output": 0 } }, - "qwen/qwen-turbo": { - "id": "qwen/qwen-turbo", - "name": "Qwen: Qwen-Turbo", - "attachment": false, + "google/gemini-2.0-flash-lite": { + "id": "google/gemini-2.0-flash-lite", + "name": "Gemini-2.0-Flash-Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-11-01", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-02-05", + "last_updated": "2025-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, + "context": 990000, "output": 8192 }, "cost": { - "input": 0.0325, - "output": 0.13, - "cache_read": 0.01 + "input": 0.052, + "output": 0.21 } }, - "qwen/qwen3-vl-235b-a22b-instruct": { - "id": "qwen/qwen3-vl-235b-a22b-instruct", - "name": "Qwen: Qwen3 VL 235B A22B Instruct", + "elevenlabs/elevenlabs-music": { + "id": "elevenlabs/elevenlabs-music", + "name": "ElevenLabs-Music", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "elevenlabs", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-23", - "last_updated": "2026-01-10", + "temperature": false, + "release_date": "2025-08-29", + "last_updated": "2025-08-29", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text"], + "output": ["audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 52429 - }, - "cost": { - "input": 0.2, - "output": 0.88, - "cache_read": 0.11 + "context": 2000, + "output": 0 } }, - "qwen/qwen3-coder": { - "id": "qwen/qwen3-coder", - "name": "Qwen: Qwen3 Coder 480B A35B", - "attachment": false, + "elevenlabs/elevenlabs-v2.5-turbo": { + "id": "elevenlabs/elevenlabs-v2.5-turbo", + "name": "ElevenLabs-v2.5-Turbo", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "elevenlabs", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "temperature": false, + "release_date": "2024-10-28", + "last_updated": "2024-10-28", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 52429 - }, - "cost": { - "input": 0.22, - "output": 1, - "cache_read": 0.022 + "context": 128000, + "output": 0 } }, - "qwen/qwen3.5-9b": { - "id": "qwen/qwen3.5-9b", - "name": "Qwen: Qwen3.5-9B", + "elevenlabs/elevenlabs-v3": { + "id": "elevenlabs/elevenlabs-v3", + "name": "ElevenLabs-v3", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "elevenlabs", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-03-10", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["image", "text", "video"], - "output": ["text"] + "input": ["text"], + "output": ["audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 32768 - }, - "cost": { - "input": 0.05, - "output": 0.15 + "context": 128000, + "output": 0 } }, - "qwen/qwen3-vl-8b-thinking": { - "id": "qwen/qwen3-vl-8b-thinking", - "name": "Qwen: Qwen3 VL 8B Thinking", + "openai/chatgpt-4o-latest": { + "id": "openai/chatgpt-4o-latest", + "name": "ChatGPT-4o-Latest", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "temperature": false, + "release_date": "2024-08-14", + "last_updated": "2024-08-14", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.117, - "output": 1.365 + "input": 4.5, + "output": 14 } }, - "qwen/qwen3.6-max-preview": { - "id": "qwen/qwen3.6-max-preview", - "name": "Qwen: Qwen3.6 Max Preview", - "attachment": false, - "reasoning": true, + "openai/gpt-3.5-turbo-instruct": { + "id": "openai/gpt-3.5-turbo-instruct", + "name": "GPT-3.5-Turbo-Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": false, + "release_date": "2023-09-20", + "last_updated": "2023-09-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 3500, + "output": 1024 }, "cost": { - "input": 1.04, - "output": 6.24, - "cache_write": 1.3 + "input": 1.4, + "output": 1.8 } }, - "qwen/qwen-plus-2025-07-28:thinking": { - "id": "qwen/qwen-plus-2025-07-28:thinking", - "name": "Qwen: Qwen Plus 0728 (thinking)", - "attachment": false, + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-09-09", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.26, - "output": 0.78 + "input": 1.8, + "output": 7.2, + "cache_read": 0.45 } }, - "qwen/qwen-2.5-72b-instruct": { - "id": "qwen/qwen-2.5-72b-instruct", - "name": "Qwen2.5 72B Instruct", - "attachment": false, + "openai/gpt-5.2-instant": { + "id": "openai/gpt-5.2-instant", + "name": "GPT-5.2-Instant", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-09", - "last_updated": "2026-01-10", + "temperature": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.12, - "output": 0.39 + "input": 1.6, + "output": 13, + "cache_read": 0.16 } }, - "qwen/qwen3-14b": { - "id": "qwen/qwen3-14b", - "name": "Qwen: Qwen3 14B", - "attachment": false, - "reasoning": true, + "openai/gpt-4o-search": { + "id": "openai/gpt-4o-search", + "name": "GPT-4o-Search", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-04", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-03-11", + "last_updated": "2025-03-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.06, - "output": 0.24, - "cache_read": 0.025 + "input": 2.2, + "output": 9 } }, - "qwen/qwen3.5-35b-a3b": { - "id": "qwen/qwen3.5-35b-a3b", - "name": "Qwen: Qwen3.5-35B-A3B", + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT-5.2-Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.1625, - "output": 1.3 + "input": 19, + "output": 150 } }, - "qwen/qwen3.5-plus-02-15": { - "id": "qwen/qwen3.5-plus-02-15", - "name": "Qwen: Qwen3.5 Plus 2026-02-15", + "openai/gpt-4-classic-0314": { + "id": "openai/gpt-4-classic-0314", + "name": "GPT-4-Classic-0314", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-08-26", + "last_updated": "2024-08-26", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 8192, + "output": 4096 }, + "status": "deprecated", "cost": { - "input": 0.26, - "output": 1.56 + "input": 27, + "output": 54 } }, - "qwen/qwen3.6-flash": { - "id": "qwen/qwen3.6-flash", - "name": "Qwen: Qwen3.6 Flash", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_write": 0.3125 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "alfredpros/codellama-7b-instruct-solidity": { - "id": "alfredpros/codellama-7b-instruct-solidity", - "name": "AlfredPros: CodeLLaMa 7B Instruct Solidity", - "attachment": false, + "openai/gpt-5-chat": { + "id": "openai/gpt-5-chat", + "name": "GPT-5-Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.8, - "output": 1.2 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "kwaipilot/kat-coder-pro-v2": { - "id": "kwaipilot/kat-coder-pro-v2", - "name": "Kwaipilot: KAT-Coder-Pro V2", - "attachment": false, + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5-Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-04-11", + "temperature": false, + "release_date": "2023-09-13", + "last_updated": "2023-09-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 80000 + "context": 16384, + "output": 2048 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0.45, + "output": 1.4 } }, - "google/gemini-2.5-pro-preview-05-06": { - "id": "google/gemini-2.5-pro-preview-05-06", - "name": "Google: Gemini 2.5 Pro Preview 05-06", + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT-5-Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-05-06", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "reasoning": 10, - "cache_read": 0.125, - "cache_write": 0.375 + "input": 14, + "output": 110 } }, - "google/lyria-3-clip-preview": { - "id": "google/lyria-3-clip-preview", - "name": "Google: Lyria 3 Clip Preview", + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-04-11", + "tool_call": true, + "temperature": false, + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { - "input": ["image", "text"], - "output": ["audio", "text"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 - }, - "cost": { - "input": 0, - "output": 0 + "context": 128000, + "output": 8192 } }, - "google/gemini-3.1-pro-preview-customtools": { - "id": "google/gemini-3.1-pro-preview-customtools", - "name": "Google: Gemini 3.1 Pro Preview Custom Tools", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 2, - "output": 12, - "reasoning": 12 + "input": 0.99, + "output": 4, + "cache_read": 0.25 } }, - "google/gemini-2.5-flash-lite-preview-09-2025": { - "id": "google/gemini-2.5-flash-lite-preview-09-2025", - "name": "Google: Gemini 2.5 Flash Lite Preview 09-2025", + "openai/sora-2": { + "id": "openai/sora-2", + "name": "Sora-2", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "sora", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 - }, - "cost": { - "input": 0.1, - "output": 0.4, - "reasoning": 0.4, - "cache_read": 0.01, - "cache_write": 0.083333 + "context": 0, + "output": 0 } }, - "google/gemini-2.0-flash-001": { - "id": "google/gemini-2.0-flash-001", - "name": "Google: Gemini 2.0 Flash", + "openai/o3-pro": { + "id": "openai/o3-pro", + "name": "o3-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-12-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-06-10", + "last_updated": "2025-06-10", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025, - "cache_write": 0.083333 + "input": 18, + "output": 72 } }, - "google/lyria-3-pro-preview": { - "id": "google/lyria-3-pro-preview", - "name": "Google: Lyria 3 Pro Preview", + "openai/gpt-4-classic": { + "id": "openai/gpt-4-classic", + "name": "GPT-4-Classic", + "description": "Legacy model retained for compatibility with older integrations", + "family": "gpt", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-04-11", + "tool_call": true, + "temperature": false, + "release_date": "2024-03-25", + "last_updated": "2024-03-25", "modalities": { - "input": ["image", "text"], - "output": ["audio", "text"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 8192, + "output": 4096 }, + "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 27, + "output": 54 } }, - "google/gemma-3n-e4b-it": { - "id": "google/gemma-3n-e4b-it", - "name": "Google: Gemma 3n 4B", - "attachment": false, + "openai/gpt-4o-aug": { + "id": "openai/gpt-4o-aug", + "name": "GPT-4o-Aug", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "tool_call": true, + "temperature": false, + "release_date": "2024-11-21", + "last_updated": "2024-11-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 6554 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.02, - "output": 0.04 + "input": 2.2, + "output": 9, + "cache_read": 1.1 } }, - "google/gemini-3.1-flash-lite-preview": { - "id": "google/gemini-3.1-flash-lite-preview", - "name": "Google: Gemini 3.1 Flash Lite Preview", + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4-Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-03", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 1.5, - "reasoning": 1.5 + "input": 0.18, + "output": 1.1, + "cache_read": 0.018 } }, - "google/gemini-3.1-pro-preview": { - "id": "google/gemini-3.1-pro-preview", - "name": "Google: Gemini 3.1 Pro Preview", + "openai/sora-2-pro": { + "id": "openai/sora-2-pro", + "name": "Sora-2-Pro", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "sora", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-02-19", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 - }, - "cost": { - "input": 2, - "output": 12, - "reasoning": 12 + "context": 0, + "output": 0 } }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Google: Gemini 3 Flash Preview", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-12-17", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-11-12", + "last_updated": "2025-11-12", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 3, - "reasoning": 3, - "cache_read": 0.05, - "cache_write": 0.083333 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Google: Gemini 2.5 Pro", + "openai/gpt-5.3-instant": { + "id": "openai/gpt-5.3-instant", + "name": "GPT-5.3-Instant", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-03-20", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "input": 111616, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "reasoning": 10, - "cache_read": 0.125, - "cache_write": 0.375 + "input": 1.6, + "output": 13, + "cache_read": 0.16 } }, - "google/gemini-3-pro-image-preview": { - "id": "google/gemini-3-pro-image-preview", - "name": "Google: Nano Banana Pro (Gemini 3 Pro Image Preview)", + "openai/gpt-5.3-codex-spark": { + "id": "openai/gpt-5.3-codex-spark", + "name": "GPT-5.3-Codex-Spark", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-11-20", - "last_updated": "2026-03-15", + "reasoning_options": [], + "tool_call": true, + "temperature": false, + "release_date": "2026-03-04", + "last_updated": "2026-03-04", "modalities": { - "input": ["image", "text"], - "output": ["image", "text"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 12, - "reasoning": 12 + "input": 0, + "output": 0 } }, - "google/gemma-4-31b-it": { - "id": "google/gemma-4-31b-it", - "name": "Google: Gemma 4 31B", + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT-5.1-Codex-Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-11", + "temperature": false, + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.4 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "google/gemini-2.5-flash-image": { - "id": "google/gemini-2.5-flash-image", - "name": "Google: Nano Banana (Gemini 2.5 Flash Image)", + "openai/dall-e-3": { + "id": "openai/dall-e-3", + "name": "DALL-E-3", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "dall-e", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-10-08", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2023-11-06", + "last_updated": "2023-11-06", "modalities": { - "input": ["image", "text"], - "output": ["image", "text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 - }, - "cost": { - "input": 0.3, - "output": 2.5 + "context": 800, + "output": 0 } }, - "google/gemma-3-12b-it": { - "id": "google/gemma-3-12b-it", - "name": "Google: Gemma 3 12B", + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-03-13", - "last_updated": "2026-03-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.04, - "output": 0.13, - "cache_read": 0.015 + "input": 0.99, + "output": 4 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Google: Gemini 2.5 Flash", + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-17", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 2.5, - "reasoning": 2.5, - "cache_read": 0.03, - "cache_write": 0.083333 + "input": 1.6, + "output": 13, + "cache_read": 0.16 } }, - "google/gemini-3.1-flash-image-preview": { - "id": "google/gemini-3.1-flash-image-preview", - "name": "Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)", + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2026-02-10", + "last_updated": "2026-02-10", "modalities": { - "input": ["image", "text"], - "output": ["image", "text"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 3 + "input": 1.6, + "output": 13, + "cache_read": 0.16 } }, - "google/gemma-3-4b-it": { - "id": "google/gemma-3-4b-it", - "name": "Google: Gemma 3 4B", + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1-Codex-Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-03-13", - "last_updated": "2026-03-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-11-12", + "last_updated": "2025-11-12", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 19200 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.04, - "output": 0.08 + "input": 0.22, + "output": 1.8, + "cache_read": 0.022 } }, - "google/gemini-2.5-pro-preview": { - "id": "google/gemini-2.5-pro-preview", - "name": "Google: Gemini 2.5 Pro Preview 06-05", + "openai/o4-mini-deep-research": { + "id": "openai/o4-mini-deep-research", + "name": "o4-mini-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o-mini", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-06-05", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-06-27", + "last_updated": "2025-06-27", "modalities": { - "input": ["audio", "image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 1.25, - "output": 10, - "reasoning": 10, - "cache_read": 0.125, - "cache_write": 0.375 + "input": 1.8, + "output": 7.2, + "cache_read": 0.45 } }, - "google/gemma-2-27b-it": { - "id": "google/gemma-2-27b-it", - "name": "Google: Gemma 2 27B", - "attachment": false, + "openai/gpt-image-1.5": { + "id": "openai/gpt-image-1.5", + "name": "gpt-image-1.5", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2024-06-24", - "last_updated": "2024-06-24", + "temperature": false, + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 - }, - "cost": { - "input": 0.65, - "output": 0.65 + "context": 128000, + "output": 0 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Google: Gemma 3 27B", + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-03-12", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["image", "text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.03, - "output": 0.11, - "cache_read": 0.02 + "input": 0.09, + "output": 0.36, + "cache_read": 0.022 } }, - "google/gemma-4-26b-a4b-it": { - "id": "google/gemma-4-26b-a4b-it", - "name": "Google: Gemma 4 26B A4B", + "openai/gpt-3.5-turbo-raw": { + "id": "openai/gpt-3.5-turbo-raw", + "name": "GPT-3.5-Turbo-Raw", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2026-04-03", - "last_updated": "2026-04-11", + "temperature": false, + "release_date": "2023-09-27", + "last_updated": "2023-09-27", "modalities": { - "input": ["image", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 4524, + "output": 2048 }, "cost": { - "input": 0.12, - "output": 0.4 + "input": 0.45, + "output": 1.4 } }, - "google/gemini-2.5-flash-lite": { - "id": "google/gemini-2.5-flash-lite", - "name": "Google: Gemini 2.5 Flash Lite", + "openai/o1": { + "id": "openai/o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-06-17", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2024-12-18", + "last_updated": "2024-12-18", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.1, - "output": 0.4, - "reasoning": 0.4, - "cache_read": 0.01, - "cache_write": 0.083333 + "input": 14, + "output": 54 } }, - "google/gemini-2.0-flash-lite-001": { - "id": "google/gemini-2.0-flash-lite-001", - "name": "Google: Gemini 2.0 Flash Lite", + "openai/o1-pro": { + "id": "openai/o1-pro", + "name": "o1-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-12-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-03-19", + "last_updated": "2025-03-19", "modalities": { - "input": ["audio", "image", "pdf", "text", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 140, + "output": 540 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "MoonshotAI: Kimi K2.5", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65535 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 2.2 + "input": 2.2, + "output": 14, + "cache_read": 0.22 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "MoonshotAI: Kimi K2 0905", - "attachment": false, - "reasoning": false, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4-Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "temperature": false, + "release_date": "2026-03-12", + "last_updated": "2026-03-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 26215 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.15 + "input": 0.68, + "output": 4, + "cache_read": 0.068 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "MoonshotAI: Kimi K2.6", + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-05-12", + "temperature": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65535 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.75, - "output": 3.5, - "cache_read": 0.375 + "input": 1.8, + "output": 7.2, + "cache_read": 0.45 } }, - "moonshotai/kimi-k2": { - "id": "moonshotai/kimi-k2", - "name": "MoonshotAI: Kimi K2 0711", - "attachment": false, - "reasoning": false, + "openai/o3-deep-research": { + "id": "openai/o3-deep-research", + "name": "o3-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-07-11", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-06-27", + "last_updated": "2025-06-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131000, - "output": 26215 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 9, + "output": 36, + "cache_read": 2.2 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "MoonshotAI: Kimi K2 Thinking", - "attachment": false, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-11-06", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-06-25", + "last_updated": "2025-06-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65535 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.47, - "output": 2, - "cache_read": 0.2 + "input": 0.22, + "output": 1.8, + "cache_read": 0.022 } }, - "aion-labs/aion-1.0": { - "id": "aion-labs/aion-1.0", - "name": "AionLabs: Aion-1.0", - "attachment": false, - "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-02-05", - "last_updated": "2026-03-15", + "openai/gpt-image-1": { + "id": "openai/gpt-image-1", + "name": "GPT-Image-1", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-03-31", + "last_updated": "2025-03-31", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 0 + } + }, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, + "context": 1047576, "output": 32768 }, "cost": { - "input": 4, - "output": 8 + "input": 0.36, + "output": 1.4, + "cache_read": 0.09 } }, - "aion-labs/aion-rp-llama-3.1-8b": { - "id": "aion-labs/aion-rp-llama-3.1-8b", - "name": "AionLabs: Aion-RP 1.0 (8B)", - "attachment": false, + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4-Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-02-05", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2023-09-13", + "last_updated": "2023-09-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.8, - "output": 1.6 + "input": 9, + "output": 27 } }, - "aion-labs/aion-2.0": { - "id": "aion-labs/aion-2.0", - "name": "AionLabs: Aion-2.0", - "attachment": false, + "openai/gpt-image-1-mini": { + "id": "openai/gpt-image-1-mini", + "name": "GPT-Image-1-Mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-08-26", + "last_updated": "2025-08-26", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + } + }, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.8, - "output": 1.6 + "input": 0.045, + "output": 0.36, + "cache_read": 0.0045 } }, - "aion-labs/aion-1.0-mini": { - "id": "aion-labs/aion-1.0-mini", - "name": "AionLabs: Aion-1.0-Mini", - "attachment": false, + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT-5.4-Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-02-05", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.7, - "output": 1.4 + "input": 27, + "output": 160 } }, - "~moonshotai/kimi-latest": { - "id": "~moonshotai/kimi-latest", - "name": "MoonshotAI: Kimi Latest", + "openai/o3-mini-high": { + "id": "openai/o3-mini-high", + "name": "o3-mini-high", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-04-27", - "last_updated": "2026-05-01", + "temperature": false, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262142, - "output": 262142 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.74, - "output": 3.49, - "cache_read": 0.14 + "input": 0.99, + "output": 4 } }, - "thedrummer/unslopnemo-12b": { - "id": "thedrummer/unslopnemo-12b", - "name": "TheDrummer: UnslopNemo 12B", - "attachment": false, - "reasoning": false, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5-Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-11-09", - "last_updated": "2026-03-15", + "structured_output": false, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-08", + "last_updated": "2026-04-08", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 27.2727, + "output": 163.6364 } }, - "thedrummer/cydonia-24b-v4.1": { - "id": "thedrummer/cydonia-24b-v4.1", - "name": "TheDrummer: Cydonia 24B V4.1", - "attachment": false, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-09-27", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 124096, + "output": 4096 }, "cost": { - "input": 0.3, - "output": 0.5 + "input": 0.14, + "output": 0.54, + "cache_read": 0.068 } }, - "thedrummer/skyfall-36b-v2": { - "id": "thedrummer/skyfall-36b-v2", - "name": "TheDrummer: Skyfall 36B V2", - "attachment": false, + "openai/gpt-4o-mini-search": { + "id": "openai/gpt-4o-mini-search", + "name": "GPT-4o-mini-Search", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, + "tool_call": true, + "temperature": false, "release_date": "2025-03-11", - "last_updated": "2026-03-15", + "last_updated": "2025-03-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.55, - "output": 0.8 + "input": 0.14, + "output": 0.54 } }, - "thedrummer/rocinante-12b": { - "id": "thedrummer/rocinante-12b", - "name": "TheDrummer: Rocinante 12B", - "attachment": false, - "reasoning": false, + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2024-09-30", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.17, - "output": 0.43 + "input": 1.1, + "output": 9 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Anthropic: Claude Opus 4.1", + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1.6, + "output": 13, + "cache_read": 0.16 } }, - "anthropic/claude-3.7-sonnet:thinking": { - "id": "anthropic/claude-3.7-sonnet:thinking", - "name": "Anthropic: Claude 3.7 Sonnet (thinking)", + "openai/gpt-5.1-instant": { + "id": "openai/gpt-5.1-instant", + "name": "GPT-5.1-Instant", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-02-19", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-11-12", + "last_updated": "2025-11-12", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "anthropic/claude-opus-4.6-fast": { - "id": "anthropic/claude-opus-4.6-fast", - "name": "Anthropic: Claude Opus 4.6 (Fast)", + "openai/gpt-image-2": { + "id": "openai/gpt-image-2", + "name": "GPT-Image-2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-04-07", - "last_updated": "2026-04-11", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["image", "text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 0, + "output": 0 }, "cost": { - "input": 30, - "output": 150, - "cache_read": 3, - "cache_write": 37.5 + "input": 5.0505, + "output": 32.3232, + "cache_read": 1.2626 } }, - "anthropic/claude-3.7-sonnet": { - "id": "anthropic/claude-3.7-sonnet", - "name": "Anthropic: Claude 3.7 Sonnet", + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-02-19", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-11-12", + "last_updated": "2025-11-12", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.1, + "output": 9, + "cache_read": 0.11 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Anthropic: Claude Opus 4.6", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-08", + "last_updated": "2026-04-08", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 4.5455, + "output": 27.2727, + "cache_read": 0.4545 } }, - "anthropic/claude-opus-4.7": { - "id": "anthropic/claude-opus-4.7", - "name": "Anthropic: Claude Opus 4.7", + "runwayml/runway": { + "id": "runwayml/runway", + "name": "Runway", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "runway", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": false, - "release_date": "2026-04-16", - "last_updated": "2026-05-01", + "release_date": "2024-10-11", + "last_updated": "2024-10-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], + "output": ["video"] + }, + "open_weights": false, + "limit": { + "context": 256, + "output": 0 + } + }, + "runwayml/runway-gen-4-turbo": { + "id": "runwayml/runway-gen-4-turbo", + "name": "Runway-Gen-4-Turbo", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "runway", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-05-09", + "last_updated": "2025-05-09", + "modalities": { + "input": ["text", "image"], + "output": ["video"] + }, + "open_weights": false, + "limit": { + "context": 256, + "output": 0 + } + }, + "cerebras/llama-3.1-8b-cs": { + "id": "cerebras/llama-3.1-8b-cs", + "name": "Llama-3.1-8B-CS", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": false, + "release_date": "2025-05-13", + "last_updated": "2025-05-13", + "modalities": { + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "output": 0 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.1, + "output": 0.1 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Anthropic: Claude Sonnet 4", + "cerebras/qwen3-235b-2507-cs": { + "id": "cerebras/qwen3-235b-2507-cs", + "name": "qwen3-235b-2507-cs", + "description": "Legacy model retained for compatibility with older integrations", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-05-22", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 0, + "output": 0 }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 - } + "status": "deprecated" }, - "anthropic/claude-sonnet-4.5": { - "id": "anthropic/claude-sonnet-4.5", - "name": "Anthropic: Claude Sonnet 4.5", + "cerebras/gpt-oss-120b-cs": { + "id": "cerebras/gpt-oss-120b-cs", + "name": "GPT-OSS-120B-CS", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-08-06", + "last_updated": "2025-08-06", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 128000, + "output": 0 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.35, + "output": 0.75 } }, - "anthropic/claude-opus-4.5": { - "id": "anthropic/claude-opus-4.5", - "name": "Anthropic: Claude Opus 4.5", + "cerebras/llama-3.3-70b-cs": { + "id": "cerebras/llama-3.3-70b-cs", + "name": "llama-3.3-70b-cs", + "description": "Legacy model retained for compatibility with older integrations", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-05-13", + "last_updated": "2025-05-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 0, + "output": 0 + }, + "status": "deprecated" + }, + "cerebras/qwen3-32b-cs": { + "id": "cerebras/qwen3-32b-cs", + "name": "qwen3-32b-cs", + "description": "Legacy model retained for compatibility with older integrations", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-11-24", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-05-15", + "last_updated": "2025-05-15", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 0, + "output": 0 }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } + "status": "deprecated" }, - "anthropic/claude-3-haiku": { - "id": "anthropic/claude-3-haiku", - "name": "Anthropic: Claude 3 Haiku", + "anthropic/claude-sonnet-3.5": { + "id": "anthropic/claude-sonnet-3.5", + "name": "Claude-Sonnet-3.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-sonnet", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-03-07", - "last_updated": "2024-03-07", + "temperature": false, + "release_date": "2024-06-05", + "last_updated": "2024-06-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 189096, + "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Anthropic: Claude Opus 4", + "anthropic/claude-sonnet-3.7": { + "id": "anthropic/claude-sonnet-3.7", + "name": "Claude-Sonnet-3.7", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-05-22", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["image", "pdf", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 196608, + "output": 128000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } }, - "anthropic/claude-3.5-haiku": { - "id": "anthropic/claude-3.5-haiku", - "name": "Anthropic: Claude 3.5 Haiku", + "anthropic/claude-sonnet-3.5-june": { + "id": "anthropic/claude-sonnet-3.5-june", + "name": "Claude-Sonnet-3.5-June", + "description": "Legacy model retained for compatibility with older integrations", + "family": "claude-sonnet", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "temperature": false, + "release_date": "2024-11-18", + "last_updated": "2024-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 189096, "output": 8192 }, + "status": "deprecated", "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } }, - "anthropic/claude-haiku-4.5": { - "id": "anthropic/claude-haiku-4.5", - "name": "Anthropic: Claude Haiku 4.5", + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Claude-Sonnet-4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 31999 + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": false, + "release_date": "2025-09-26", + "last_updated": "2025-09-26", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 983040, + "output": 32768 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Anthropic: Claude Sonnet 4.6", + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude-Sonnet-4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-15", + "temperature": false, + "release_date": "2025-05-21", + "last_updated": "2025-05-21", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 983040, + "output": 64000 }, "cost": { - "input": 3, - "output": 15 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } }, - "switchpoint/router": { - "id": "switchpoint/router", - "name": "Switchpoint Router", - "attachment": false, + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Claude-Haiku-4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2025-07-12", - "last_updated": "2026-03-15", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 63999 + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 192000, + "output": 64000 }, "cost": { "input": 0.85, - "output": 3.4 + "output": 4.3, + "cache_read": 0.085, + "cache_write": 1.1 } }, - "bytedance/ui-tars-1.5-7b": { - "id": "bytedance/ui-tars-1.5-7b", - "name": "ByteDance: UI-TARS 7B ", + "anthropic/claude-haiku-3": { + "id": "anthropic/claude-haiku-3", + "name": "Claude-Haiku-3", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2026-03-15", + "tool_call": true, + "temperature": false, + "release_date": "2024-03-09", + "last_updated": "2024-03-09", "modalities": { - "input": ["image", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 2048 + "context": 189096, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.2 + "input": 0.21, + "output": 1.1, + "cache_read": 0.021, + "cache_write": 0.26 } }, - "tngtech/deepseek-r1t2-chimera": { - "id": "tngtech/deepseek-r1t2-chimera", - "name": "TNG: DeepSeek R1T2 Chimera", - "attachment": false, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude-Opus-4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", + "temperature": false, + "release_date": "2026-04-15", + "last_updated": "2026-04-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 1048576, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 0.85, - "cache_read": 0.125 + "input": 4.3, + "output": 21, + "cache_read": 0.43, + "cache_write": 5.4 } }, - "xiaomi/mimo-v2.5-pro": { - "id": "xiaomi/mimo-v2.5-pro", - "name": "Xiaomi: MiMo V2.5 Pro", - "family": "mimo", - "attachment": false, - "reasoning": true, + "anthropic/claude-haiku-3.5": { + "id": "anthropic/claude-haiku-3.5", + "name": "Claude-Haiku-3.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "temperature": false, + "release_date": "2024-10-01", + "last_updated": "2024-10-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 189096, + "output": 8192 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.68, + "output": 3.4, + "cache_read": 0.068, + "cache_write": 0.85 } }, - "xiaomi/mimo-v2-omni": { - "id": "xiaomi/mimo-v2-omni", - "name": "Xiaomi: MiMo-V2-Omni", - "family": "mimo", + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude-Opus-4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1048576, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 4.2929, + "output": 21.4646 } }, - "xiaomi/mimo-v2.5": { - "id": "xiaomi/mimo-v2.5", - "name": "Xiaomi: MiMo-V2.5", - "family": "mimo", + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude-Opus-4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 31999 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "temperature": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 196608, + "output": 32000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - }, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 13, + "output": 64, + "cache_read": 1.3, + "cache_write": 16 } }, - "xiaomi/mimo-v2-pro": { - "id": "xiaomi/mimo-v2-pro", - "name": "Xiaomi: MiMo-V2-Pro", - "family": "mimo", - "attachment": false, + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Claude-Opus-4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 0, + "max": 63999 + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "temperature": false, + "release_date": "2025-11-21", + "last_updated": "2025-11-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 196608, + "output": 64000 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 4.3, + "output": 21, + "cache_read": 0.43, + "cache_write": 5.3 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "Xiaomi: MiMo-V2-Flash", - "family": "mimo", - "attachment": false, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude-Sonnet-4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "temperature": false, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 983040, + "output": 128000 }, "cost": { - "input": 0.09, - "output": 0.29, - "cache_read": 0.045 + "input": 2.6, + "output": 13, + "cache_read": 0.26, + "cache_write": 3.2 } - } - } - }, - "sap-ai-core": { - "id": "sap-ai-core", - "env": ["AICORE_SERVICE_KEY"], - "npm": "@jerome-benoit/sap-ai-provider-v2", - "name": "SAP AI Core", - "doc": "https://help.sap.com/docs/sap-ai-core", - "models": { - "anthropic--claude-4.6-opus": { - "id": "anthropic--claude-4.6-opus", - "name": "anthropic--claude-4.6-opus", + }, + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude-Opus-4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "temperature": false, + "release_date": "2025-05-21", + "last_updated": "2025-05-21", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 192512, + "output": 28672 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 13, + "output": 64, + "cache_read": 1.3, + "cache_write": 16 } }, - "anthropic--claude-3-haiku": { - "id": "anthropic--claude-3-haiku", - "name": "anthropic--claude-3-haiku", - "family": "claude-haiku", + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude-Opus-4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-13", - "last_updated": "2024-03-13", + "temperature": false, + "release_date": "2026-02-04", + "last_updated": "2026-02-04", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 983040, + "output": 128000 }, "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 4.3, + "output": 21, + "cache_read": 0.43, + "cache_write": 5.3 } }, - "anthropic--claude-3-opus": { - "id": "anthropic--claude-3-opus", - "name": "anthropic--claude-3-opus", - "family": "claude-opus", + "lumalabs/ray2": { + "id": "lumalabs/ray2", + "name": "Ray2", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "ray", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-02-29", - "last_updated": "2024-02-29", + "temperature": false, + "release_date": "2025-02-20", + "last_updated": "2025-02-20", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 - }, - "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "context": 5000, + "output": 0 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "gpt-5-mini", - "family": "gpt-mini", + "poetools/claude-code": { + "id": "poetools/claude-code", + "name": "claude-code", + "description": "Claude model for careful reasoning, writing, coding, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2025-11-27", + "last_updated": "2025-11-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "context": 0, + "output": 0 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "gpt-5-nano", - "family": "gpt-nano", + "empiriolabs/deepseek-v4-pro-el": { + "id": "empiriolabs/deepseek-v4-pro-el", + "name": "DeepSeek-V4-Pro-EL", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2026-04-24", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "input": 1000000, + "output": 384000 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 1.67, + "output": 3.33 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "gemini-2.5-pro", - "family": "gemini-pro", + "empiriolabs/deepseek-v4-flash-el": { + "id": "empiriolabs/deepseek-v4-flash-el", + "name": "DeepSeek-V4-Flash-EL", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-25", - "last_updated": "2025-06-05", + "release_date": "2026-04-24", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "input": 1000000, + "output": 384000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.14, + "output": 0.28 } }, - "anthropic--claude-3.7-sonnet": { - "id": "anthropic--claude-3.7-sonnet", - "name": "anthropic--claude-3.7-sonnet", - "family": "claude-sonnet", + "stabilityai/stablediffusionxl": { + "id": "stabilityai/stablediffusionxl", + "name": "StableDiffusionXL", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "stable-diffusion", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "temperature": false, + "release_date": "2023-07-09", + "last_updated": "2023-07-09", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "context": 200, + "output": 0 } }, - "sonar-pro": { - "id": "sonar-pro", - "name": "sonar-pro", - "family": "sonar-pro", + "ideogramai/ideogram": { + "id": "ideogramai/ideogram", + "name": "Ideogram", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ideogram", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "tool_call": true, + "temperature": false, + "release_date": "2024-04-03", + "last_updated": "2024-04-03", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 - }, - "cost": { - "input": 3, - "output": 15 + "context": 150, + "output": 0 } }, - "anthropic--claude-4.5-sonnet": { - "id": "anthropic--claude-4.5-sonnet", - "name": "anthropic--claude-4.5-sonnet", - "family": "claude-sonnet", + "ideogramai/ideogram-v2a": { + "id": "ideogramai/ideogram-v2a", + "name": "Ideogram-v2a", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ideogram", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "temperature": false, + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "context": 150, + "output": 0 } }, - "anthropic--claude-4.6-sonnet": { - "id": "anthropic--claude-4.6-sonnet", - "name": "anthropic--claude-4.6-sonnet", - "family": "claude-sonnet", + "ideogramai/ideogram-v2": { + "id": "ideogramai/ideogram-v2", + "name": "Ideogram-v2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ideogram", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-08", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "temperature": false, + "release_date": "2024-08-21", + "last_updated": "2024-08-21", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "context": 150, + "output": 0 } }, - "sonar-deep-research": { - "id": "sonar-deep-research", - "name": "sonar-deep-research", - "family": "sonar-deep-research", - "attachment": false, - "reasoning": true, - "tool_call": false, + "ideogramai/ideogram-v2a-turbo": { + "id": "ideogramai/ideogram-v2a-turbo", + "name": "Ideogram-v2a-Turbo", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "ideogram", + "attachment": true, + "reasoning": false, + "tool_call": true, "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-02-01", - "last_updated": "2025-09-01", + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 - }, - "cost": { - "input": 2, - "output": 8, - "reasoning": 3 + "context": 150, + "output": 0 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "gemini-2.5-flash", - "family": "gemini-flash", + "novita/glm-4.7": { + "id": "novita/glm-4.7", + "name": "glm-4.7", + "description": "Legacy model retained for compatibility with older integrations", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-25", - "last_updated": "2025-06-05", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 205000, + "output": 131072 }, - "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "input_audio": 1 - } + "status": "deprecated" }, - "anthropic--claude-4.5-opus": { - "id": "anthropic--claude-4.5-opus", - "name": "anthropic--claude-4.5-opus", - "family": "claude-opus", + "novita/minimax-m2.1": { + "id": "novita/minimax-m2.1", + "name": "minimax-m2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "temperature": false, + "release_date": "2025-12-26", + "last_updated": "2025-12-26", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "context": 205000, + "output": 131072 } }, - "sonar": { - "id": "sonar", - "name": "sonar", - "family": "sonar", - "attachment": false, + "novita/glm-4.6": { + "id": "novita/glm-4.6", + "name": "GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "tool_call": true, + "temperature": false, + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 - }, - "cost": { - "input": 1, - "output": 1 + "context": 0, + "output": 0 } }, - "anthropic--claude-4-opus": { - "id": "anthropic--claude-4-opus", - "name": "anthropic--claude-4-opus", - "family": "claude-opus", + "novita/kimi-k2-thinking": { + "id": "novita/kimi-k2-thinking", + "name": "kimi-k2-thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "release_date": "2025-11-07", + "last_updated": "2025-11-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 - }, - "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "context": 256000, + "output": 0 } }, - "anthropic--claude-3-sonnet": { - "id": "anthropic--claude-3-sonnet", - "name": "anthropic--claude-3-sonnet", - "family": "claude-sonnet", + "novita/kimi-k2.5": { + "id": "novita/kimi-k2.5", + "name": "Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-04", - "last_updated": "2024-03-04", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 128000, + "output": 262144 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "anthropic--claude-4-sonnet": { - "id": "anthropic--claude-4-sonnet", - "name": "anthropic--claude-4-sonnet", - "family": "claude-sonnet", + "novita/glm-4.6v": { + "id": "novita/glm-4.6v", + "name": "glm-4.6v", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": false, + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "context": 131000, + "output": 32768 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "gemini-2.5-flash-lite", - "family": "gemini-flash-lite", + "novita/kimi-k2.6": { + "id": "novita/kimi-k2.6", + "name": "Kimi-K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "input": 262144, + "output": 262144 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.96, + "output": 4.04, + "cache_read": 0.16 } }, - "anthropic--claude-4.5-haiku": { - "id": "anthropic--claude-4.5-haiku", - "name": "anthropic--claude-4.5-haiku", - "family": "claude-haiku", + "novita/glm-4.7-flash": { + "id": "novita/glm-4.7-flash", + "name": "glm-4.7-flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": false, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 64000 - }, - "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "output": 65500 } }, - "gpt-5": { - "id": "gpt-5", - "name": "gpt-5", - "family": "gpt", + "novita/glm-4.7-n": { + "id": "novita/glm-4.7-n", + "name": "glm-4.7-n", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 - }, - "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "context": 205000, + "output": 131072 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "gpt-4.1", - "family": "gpt", + "novita/glm-5": { + "id": "novita/glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 205000, + "output": 131072 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "gpt-4.1-mini", - "family": "gpt-mini", + "novita/deepseek-v3.2": { + "id": "novita/deepseek-v3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 128000, + "output": 0 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.27, + "output": 0.4, + "cache_read": 0.13 } - }, - "anthropic--claude-3.5-sonnet": { - "id": "anthropic--claude-3.5-sonnet", - "name": "anthropic--claude-3.5-sonnet", - "family": "claude-sonnet", + } + } + }, + "kimi-for-coding": { + "id": "kimi-for-coding", + "env": ["KIMI_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.kimi.com/coding/v1", + "name": "Kimi For Coding", + "doc": "https://www.kimi.com/code/docs/en/third-party-tools/other-coding-agents.html", + "models": { + "k2p7": { + "id": "k2p7", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - } - } - }, - "morph": { - "id": "morph", - "env": ["MORPH_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.morphllm.com/v1", - "name": "Morph", - "doc": "https://docs.morphllm.com/api-reference/introduction", - "models": { - "auto": { - "id": "auto", - "name": "Auto", - "family": "auto", + }, + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-11", + "last_updated": "2025-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 32000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.85, - "output": 1.55 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "morph-v3-fast": { - "id": "morph-v3-fast", - "name": "Morph v3 Fast", - "family": "morph", + "k2p5": { + "id": "k2p5", + "name": "Kimi K2.5", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16000, - "output": 16000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.8, - "output": 1.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "morph-v3-large": { - "id": "morph-v3-large", - "name": "Morph v3 Large", - "family": "morph", + "k2p6": { + "id": "k2p6", + "name": "Kimi K2.6", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04", + "last_updated": "2026-04", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 32000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.9, - "output": 1.9 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } } } }, - "cloudflare-ai-gateway": { - "id": "cloudflare-ai-gateway", - "env": ["CLOUDFLARE_API_TOKEN", "CLOUDFLARE_ACCOUNT_ID", "CLOUDFLARE_GATEWAY_ID"], - "npm": "ai-gateway-provider", - "name": "Cloudflare AI Gateway", - "doc": "https://developers.cloudflare.com/ai-gateway/", + "dinference": { + "id": "dinference", + "env": ["DINFERENCE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.dinference.com/v1", + "name": "DInference", + "doc": "https://dinference.com", "models": { - "workers-ai/@cf/myshell-ai/melotts": { - "id": "workers-ai/@cf/myshell-ai/melotts", - "name": "MyShell MeloTTS", - "family": "melotts", + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.22, + "output": 0.88 } }, - "workers-ai/@cf/ibm-granite/granite-4.0-h-micro": { - "id": "workers-ai/@cf/ibm-granite/granite-4.0-h-micro", - "name": "IBM Granite 4.0 H Micro", - "family": "granite", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.017, - "output": 0.11 + "input": 0.45, + "output": 1.65 } }, - "workers-ai/@cf/huggingface/distilbert-sst-2-int8": { - "id": "workers-ai/@cf/huggingface/distilbert-sst-2-int8", - "name": "DistilBERT SST-2 INT8", - "family": "distilbert", + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.026, - "output": 0 + "input": 1.25, + "output": 3.89 } }, - "workers-ai/@cf/zai-org/glm-4.7-flash": { - "id": "workers-ai/@cf/zai-org/glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-08", + "last_updated": "2025-08", "modalities": { "input": ["text"], "output": ["text"] @@ -54521,432 +100320,671 @@ "open_weights": true, "limit": { "context": 131072, - "output": 131072 + "output": 32768 }, "cost": { - "input": 0.06, - "output": 0.4 + "input": 0.0675, + "output": 0.27 } }, - "workers-ai/@cf/pipecat-ai/smart-turn-v2": { - "id": "workers-ai/@cf/pipecat-ai/smart-turn-v2", - "name": "Pipecat Smart Turn v2", - "family": "smart-turn", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, + "open_weights": true, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 2.4 + } + } + } + }, + "perplexity-agent": { + "id": "perplexity-agent", + "env": ["PERPLEXITY_API_KEY"], + "npm": "@ai-sdk/openai", + "api": "https://api.perplexity.ai/v1", + "name": "Perplexity Agent", + "doc": "https://docs.perplexity.ai/docs/agent-api/models", + "models": { + "xai/grok-4-1-fast-non-reasoning": { + "id": "xai/grok-4-1-fast-non-reasoning", + "name": "Grok 4.1 Fast (Non-Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-07", + "release_date": "2025-11-19", + "last_updated": "2025-11-19", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct": { - "id": "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct", - "name": "Mistral Small 3.1 24B Instruct", - "family": "mistral-small", - "attachment": false, - "reasoning": false, - "tool_call": false, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.35, - "output": 0.56 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "workers-ai/@cf/facebook/bart-large-cnn": { - "id": "workers-ai/@cf/facebook/bart-large-cnn", - "name": "BART Large CNN", - "family": "bart", - "attachment": false, - "reasoning": false, - "tool_call": false, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-09", - "last_updated": "2025-04-09", + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03 } }, - "workers-ai/@cf/aisingapore/gemma-sea-lion-v4-27b-it": { - "id": "workers-ai/@cf/aisingapore/gemma-sea-lion-v4-27b-it", - "name": "Gemma SEA-LION v4 27B IT", - "family": "gemma", - "attachment": false, - "reasoning": false, - "tool_call": false, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.35, - "output": 0.56 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "workers-ai/@cf/nvidia/nemotron-3-120b-a12b": { - "id": "workers-ai/@cf/nvidia/nemotron-3-120b-a12b", - "name": "Nemotron 3 Super 120B", - "family": "nemotron", - "attachment": false, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 1048576, + "output": 65536 }, "cost": { "input": 0.5, - "output": 1.5 + "output": 3, + "cache_read": 0.05, + "tiers": [ + { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 0.5, + "output": 3, + "cache_read": 0.05 + } } }, - "workers-ai/@cf/deepseek-ai/deepseek-r1-distill-qwen-32b": { - "id": "workers-ai/@cf/deepseek-ai/deepseek-r1-distill-qwen-32b", - "name": "DeepSeek R1 Distill Qwen 32B", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 4.88 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "workers-ai/@cf/openai/gpt-oss-20b": { - "id": "workers-ai/@cf/openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.3 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "workers-ai/@cf/openai/gpt-oss-120b": { - "id": "workers-ai/@cf/openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.35, - "output": 0.75 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "workers-ai/@cf/mistral/mistral-7b-instruct-v0.1": { - "id": "workers-ai/@cf/mistral/mistral-7b-instruct-v0.1", - "name": "Mistral 7B Instruct v0.1", - "family": "mistral", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.11, - "output": 0.19 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct": { - "id": "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 0.85 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "workers-ai/@cf/meta/llama-3-8b-instruct-awq": { - "id": "workers-ai/@cf/meta/llama-3-8b-instruct-awq", - "name": "Llama 3 8B Instruct AWQ", - "family": "llama", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "Nemotron 3 Super 120B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2026-02", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 32000 }, "cost": { - "input": 0.12, - "output": 0.27 + "input": 0.25, + "output": 2.5 } }, - "workers-ai/@cf/meta/llama-guard-3-8b": { - "id": "workers-ai/@cf/meta/llama-guard-3-8b", - "name": "Llama Guard 3 8B", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-opus-4-5": { + "id": "anthropic/claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.48, - "output": 0.03 + "input": 5, + "output": 25, + "cache_read": 0.5 } }, - "workers-ai/@cf/meta/m2m100-1.2b": { - "id": "workers-ai/@cf/meta/m2m100-1.2b", - "name": "M2M100 1.2B", - "family": "m2m", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-sonnet-4-5": { + "id": "anthropic/claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.34, - "output": 0.34 + "input": 3, + "output": 15, + "cache_read": 0.3 } }, - "workers-ai/@cf/meta/llama-2-7b-chat-fp16": { - "id": "workers-ai/@cf/meta/llama-2-7b-chat-fp16", - "name": "Llama 2 7B Chat FP16", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "anthropic/claude-opus-4-7": { + "id": "anthropic/claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.56, - "output": 6.67 + "input": 5, + "output": 25, + "cache_read": 0.5 } }, - "workers-ai/@cf/meta/llama-3.2-11b-vision-instruct": { - "id": "workers-ai/@cf/meta/llama-3.2-11b-vision-instruct", - "name": "Llama 3.2 11B Vision Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-haiku-4-5": { + "id": "anthropic/claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.049, - "output": 0.68 + "input": 1, + "output": 5, + "cache_read": 0.1 } }, - "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast": { - "id": "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast", - "name": "Llama 3.3 70B Instruct FP8 Fast", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-opus-4-6": { + "id": "anthropic/claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.29, - "output": 2.25 + "input": 5, + "output": 25, + "cache_read": 0.5 } }, - "workers-ai/@cf/meta/llama-3.2-1b-instruct": { - "id": "workers-ai/@cf/meta/llama-3.2-1b-instruct", - "name": "Llama 3.2 1B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-sonnet-4-6": { + "id": "anthropic/claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.027, - "output": 0.2 + "input": 3, + "output": 15, + "cache_read": 0.3 } }, - "workers-ai/@cf/meta/llama-3.1-8b-instruct-fp8": { - "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct-fp8", - "name": "Llama 3.1 8B Instruct FP8", - "family": "llama", + "perplexity/sonar": { + "id": "perplexity/sonar", + "name": "Sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] @@ -54954,2409 +100992,2742 @@ "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.29 + "input": 0.25, + "output": 2.5, + "cache_read": 0.0625 } - }, - "workers-ai/@cf/meta/llama-3.2-3b-instruct": { - "id": "workers-ai/@cf/meta/llama-3.2-3b-instruct", - "name": "Llama 3.2 3B Instruct", - "family": "llama", + } + } + }, + "siliconflow": { + "id": "siliconflow", + "env": ["SILICONFLOW_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.siliconflow.com/v1", + "name": "SiliconFlow", + "doc": "https://cloud.siliconflow.com/models", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "moonshotai/Kimi-K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2026-04-21", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.051, - "output": 0.34 + "input": 0.77, + "output": 4, + "cache_read": 0.2 } }, - "workers-ai/@cf/meta/llama-3.1-8b-instruct-awq": { - "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct-awq", - "name": "Llama 3.1 8B Instruct AWQ", - "family": "llama", + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "moonshotai/Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.12, - "output": 0.27 + "input": 0.45, + "output": 2.25 } }, - "workers-ai/@cf/meta/llama-3-8b-instruct": { - "id": "workers-ai/@cf/meta/llama-3-8b-instruct", - "name": "Llama 3 8B Instruct", - "family": "llama", + "baidu/ERNIE-4.5-300B-A47B": { + "id": "baidu/ERNIE-4.5-300B-A47B", + "name": "baidu/ERNIE-4.5-300B-A47B", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "ernie", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2025-07-02", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 131000, + "output": 131000 }, "cost": { "input": 0.28, - "output": 0.83 + "output": 1.1 } }, - "workers-ai/@cf/meta/llama-3.1-8b-instruct": { - "id": "workers-ai/@cf/meta/llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "ByteDance-Seed/Seed-OSS-36B-Instruct": { + "id": "ByteDance-Seed/Seed-OSS-36B-Instruct", + "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "seed", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2025-09-04", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.28, - "output": 0.8299999999999998 + "input": 0.21, + "output": 0.57 } }, - "workers-ai/@cf/qwen/qwen2.5-coder-32b-instruct": { - "id": "workers-ai/@cf/qwen/qwen2.5-coder-32b-instruct", - "name": "Qwen 2.5 Coder 32B Instruct", - "family": "qwen", + "stepfun-ai/Step-3.5-Flash": { + "id": "stepfun-ai/Step-3.5-Flash", + "name": "stepfun-ai/Step-3.5-Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "family": "step", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.66, - "output": 1 + "input": 0.1, + "output": 0.3 } }, - "workers-ai/@cf/qwen/qwen3-embedding-0.6b": { - "id": "workers-ai/@cf/qwen/qwen3-embedding-0.6b", - "name": "Qwen3 Embedding 0.6B", - "family": "qwen", + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.012, - "output": 0 + "input": 0.13, + "output": 0.4 } }, - "workers-ai/@cf/qwen/qwq-32b": { - "id": "workers-ai/@cf/qwen/qwq-32b", - "name": "QwQ 32B", - "family": "qwen", + "google/gemma-4-26B-A4B-it": { + "id": "google/gemma-4-26B-A4B-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.66, - "output": 1 + "input": 0.12, + "output": 0.4 } }, - "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8": { - "id": "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8", - "name": "Qwen3 30B A3B FP8", - "family": "qwen", + "inclusionAI/Ling-flash-2.0": { + "id": "inclusionAI/Ling-flash-2.0", + "name": "inclusionAI/Ling-flash-2.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "ling", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2025-09-18", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 131000, + "output": 131000 }, "cost": { - "input": 0.051, - "output": 0.34 + "input": 0.14, + "output": 0.57 } }, - "workers-ai/@cf/google/gemma-3-12b-it": { - "id": "workers-ai/@cf/google/gemma-3-12b-it", - "name": "Gemma 3 12B IT", - "family": "gemma", + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.35, - "output": 0.56 + "input": 0.2, + "output": 1.6 } }, - "workers-ai/@cf/moonshotai/kimi-k2.5": { - "id": "workers-ai/@cf/moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, - "reasoning": true, + "Qwen/Qwen2.5-7B-Instruct": { + "id": "Qwen/Qwen2.5-7B-Instruct", + "name": "Qwen/Qwen2.5-7B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2024-09-18", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 33000, + "output": 4000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.05, + "output": 0.05 } }, - "workers-ai/@cf/moonshotai/kimi-k2.6": { - "id": "workers-ai/@cf/moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", + "Qwen/Qwen3-VL-235B-A22B-Instruct": { + "id": "Qwen/Qwen3-VL-235B-A22B-Instruct", + "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "release_date": "2025-10-04", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.3, + "output": 1.5 } }, - "workers-ai/@cf/ai4bharat/indictrans2-en-indic-1B": { - "id": "workers-ai/@cf/ai4bharat/indictrans2-en-indic-1B", - "name": "IndicTrans2 EN-Indic 1B", - "family": "indictrans", + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.34, - "output": 0.34 + "input": 0.3, + "output": 3.2 } }, - "workers-ai/@cf/pfnet/plamo-embedding-1b": { - "id": "workers-ai/@cf/pfnet/plamo-embedding-1b", - "name": "PLaMo Embedding 1B", - "family": "plamo", + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.019, - "output": 0 + "input": 0.39, + "output": 2.34 } }, - "workers-ai/@cf/baai/bge-small-en-v1.5": { - "id": "workers-ai/@cf/baai/bge-small-en-v1.5", - "name": "BGE Small EN v1.5", - "family": "bge", + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2025-07-28", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.13, + "output": 0.6 } }, - "workers-ai/@cf/baai/bge-large-en-v1.5": { - "id": "workers-ai/@cf/baai/bge-large-en-v1.5", - "name": "BGE Large EN v1.5", - "family": "bge", + "Qwen/Qwen3-Coder-480B-A35B-Instruct": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "name": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2025-07-31", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.2, - "output": 0 + "input": 0.25, + "output": 1 } }, - "workers-ai/@cf/baai/bge-reranker-base": { - "id": "workers-ai/@cf/baai/bge-reranker-base", - "name": "BGE Reranker Base", - "family": "bge", + "Qwen/Qwen3.5-122B-A10B": { + "id": "Qwen/Qwen3.5-122B-A10B", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-09", - "last_updated": "2025-04-09", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.0031, - "output": 0 + "input": 0.26, + "output": 2.08 } }, - "workers-ai/@cf/baai/bge-base-en-v1.5": { - "id": "workers-ai/@cf/baai/bge-base-en-v1.5", - "name": "BGE Base EN v1.5", - "family": "bge", + "Qwen/Qwen3-Coder-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "name": "Qwen/Qwen3-Coder-30B-A3B-Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2025-08-01", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.067, - "output": 0 + "input": 0.07, + "output": 0.28 } }, - "workers-ai/@cf/baai/bge-m3": { - "id": "workers-ai/@cf/baai/bge-m3", - "name": "BGE M3", - "family": "bge", + "Qwen/Qwen3.5-27B": { + "id": "Qwen/Qwen3.5-27B", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.012, - "output": 0 + "input": 0.25, + "output": 2 } }, - "workers-ai/@cf/deepgram/aura-2-en": { - "id": "workers-ai/@cf/deepgram/aura-2-en", - "name": "Deepgram Aura 2 (EN)", - "family": "aura", + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2025-07-30", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.09, + "output": 0.3 } }, - "workers-ai/@cf/deepgram/aura-2-es": { - "id": "workers-ai/@cf/deepgram/aura-2-es", - "name": "Deepgram Aura 2 (ES)", - "family": "aura", - "attachment": false, - "reasoning": false, - "tool_call": false, + "Qwen/Qwen3-VL-30B-A3B-Thinking": { + "id": "Qwen/Qwen3-VL-30B-A3B-Thinking", + "name": "Qwen/Qwen3-VL-30B-A3B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2025-10-11", + "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.29, + "output": 1 } }, - "workers-ai/@cf/deepgram/nova-3": { - "id": "workers-ai/@cf/deepgram/nova-3", - "name": "Deepgram Nova 3", - "family": "nova", - "attachment": false, + "Qwen/Qwen3-VL-8B-Instruct": { + "id": "Qwen/Qwen3-VL-8B-Instruct", + "name": "Qwen/Qwen3-VL-8B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2025-10-15", + "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.18, + "output": 0.68 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", + "Qwen/Qwen3-VL-235B-A22B-Thinking": { + "id": "Qwen/Qwen3-VL-235B-A22B-Thinking", + "name": "Qwen/Qwen3-VL-235B-A22B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "release_date": "2025-10-04", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "ai-gateway-provider" + "context": 262000, + "output": 262000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.45, + "output": 3.5 } }, - "openai/gpt-4-turbo": { - "id": "openai/gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", - "attachment": true, - "reasoning": false, + "Qwen/Qwen3-8B": { + "id": "Qwen/Qwen3-8B", + "name": "Qwen/Qwen3-8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2025-04-30", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131000, + "output": 131000 }, "cost": { - "input": 10, - "output": 30 + "input": 0.06, + "output": 0.06 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, - "reasoning": true, + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen/Qwen3.5-9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "structured_output": true, + "temperature": true, + "release_date": "2026-03-03", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.1, + "output": 0.15 } }, - "openai/o3-pro": { - "id": "openai/o3-pro", - "name": "o3-pro", - "family": "o-pro", - "attachment": true, + "Qwen/Qwen3-32B": { + "id": "Qwen/Qwen3-32B", + "name": "Qwen/Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-06-10", - "last_updated": "2025-06-10", + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131000, + "output": 131000 }, "cost": { - "input": 20, - "output": 80 + "input": 0.14, + "output": 0.57 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", + "Qwen/Qwen3-VL-32B-Instruct": { + "id": "Qwen/Qwen3-VL-32B-Instruct", + "name": "Qwen/Qwen3-VL-32B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2025-10-21", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 0.2, + "output": 0.6 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "o4-mini", - "family": "o-mini", - "attachment": true, + "Qwen/Qwen3-14B": { + "id": "Qwen/Qwen3-14B", + "name": "Qwen/Qwen3-14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "release_date": "2025-04-30", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131000, + "output": 131000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 0.07, + "output": 0.28 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "Qwen/Qwen2.5-72B-Instruct": { + "id": "Qwen/Qwen2.5-72B-Instruct", + "name": "Qwen/Qwen2.5-72B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "ai-gateway-provider" + "context": 33000, + "output": 4000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.59, + "output": 0.59 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "Qwen/Qwen3-VL-30B-A3B-Instruct": { + "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "name": "Qwen/Qwen3-VL-30B-A3B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "structured_output": true, + "temperature": true, + "release_date": "2025-10-05", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.29, + "output": 1 } }, - "openai/o1": { - "id": "openai/o1", - "name": "o1", - "family": "o", + "Qwen/Qwen3-VL-32B-Thinking": { + "id": "Qwen/Qwen3-VL-32B-Thinking", + "name": "Qwen/Qwen3-VL-32B-Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "temperature": true, + "release_date": "2025-10-21", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.2, + "output": 1.5 } }, - "openai/gpt-3.5-turbo": { - "id": "openai/gpt-3.5-turbo", - "name": "GPT-3.5-turbo", - "family": "gpt", + "Qwen/Qwen3.5-35B-A3B": { + "id": "Qwen/Qwen3.5-35B-A3B", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2021-09-01", - "release_date": "2023-03-01", - "last_updated": "2023-11-06", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16385, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.5, - "output": 1.5, - "cache_read": 1.25 + "input": 0.24, + "output": 1.8 } }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "o3-mini", - "family": "o-mini", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "openai/gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "temperature": true, + "release_date": "2025-08-13", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131000, + "output": 8000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.05, + "output": 0.45 } }, - "openai/gpt-4": { - "id": "openai/gpt-4", - "name": "GPT-4", - "family": "gpt", - "attachment": true, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "openai/gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, + "structured_output": true, "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2025-08-13", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 131000, + "output": 8000 }, "cost": { - "input": 30, - "output": 60 + "input": 0.04, + "output": 0.18 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, + "tencent/Hy3-preview": { + "id": "tencent/Hy3-preview", + "name": "Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "provider": { - "npm": "ai-gateway-provider" + "context": 262144, + "output": 262144 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.066, + "output": 0.26, + "cache_read": 0.029 } }, - "openai/o3": { - "id": "openai/o3", - "name": "o3", - "family": "o", - "attachment": true, + "tencent/Hunyuan-A13B-Instruct": { + "id": "tencent/Hunyuan-A13B-Instruct", + "name": "tencent/Hunyuan-A13B-Instruct", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131000, + "output": 131000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.14, + "output": 0.57 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "GPT-4o", - "family": "gpt", + "zai-org/GLM-5V-Turbo": { + "id": "zai-org/GLM-5V-Turbo", + "name": "zai-org/GLM-5V-Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 1.2, + "output": 4, + "cache_write": 0 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", - "attachment": true, + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "zai-org/GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-06-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 205000, + "output": 205000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.95, + "output": 2.55 } }, - "anthropic/claude-haiku-4-5": { - "id": "anthropic/claude-haiku-4-5", - "name": "Claude Haiku 4.5 (latest)", - "family": "claude-haiku", - "attachment": true, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1049000, + "output": 262000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 1.4, + "output": 4.4, + "cache_write": 0 } }, - "anthropic/claude-sonnet-4-6": { - "id": "anthropic/claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "zai-org/GLM-4.5-Air": { + "id": "zai-org/GLM-4.5-Air", + "name": "zai-org/GLM-4.5-Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2025-07-28", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "ai-gateway-provider" + "context": 131000, + "output": 131000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.14, + "output": 0.86 } }, - "anthropic/claude-opus-4-7": { - "id": "anthropic/claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "zai-org/GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2026-01", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } - }, - "anthropic/claude-opus-4-1": { - "id": "anthropic/claude-opus-4-1", - "name": "Claude Opus 4.1 (latest)", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-04-08", + "last_updated": "2026-04-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 205000, + "output": 205000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1.4, + "output": 4.4, + "cache_write": 0 } }, - "anthropic/claude-3-5-haiku": { - "id": "anthropic/claude-3-5-haiku", - "name": "Claude Haiku 3.5 (latest)", - "family": "claude-haiku", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-R1": { + "id": "deepseek-ai/DeepSeek-R1", + "name": "deepseek-ai/DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2025-05-28", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 164000, + "output": 164000 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0.5, + "output": 2.18 } }, - "anthropic/claude-3.5-sonnet": { - "id": "anthropic/claude-3.5-sonnet", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-V3.1-Terminus": { + "id": "deepseek-ai/DeepSeek-V3.1-Terminus", + "name": "deepseek-ai/DeepSeek-V3.1-Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2025-09-29", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 164000, + "output": 164000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.27, + "output": 1 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4 (latest)", - "family": "claude-sonnet", - "attachment": true, + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "deepseek-ai/DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2025-08-25", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 164000, + "output": 164000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.27, + "output": 1 } }, - "anthropic/claude-opus-4-5": { - "id": "anthropic/claude-opus-4-5", - "name": "Claude Opus 4.5 (latest)", - "family": "claude-opus", - "attachment": true, + "deepseek-ai/DeepSeek-V3.2-Exp": { + "id": "deepseek-ai/DeepSeek-V3.2-Exp", + "name": "deepseek-ai/DeepSeek-V3.2-Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "release_date": "2025-10-10", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 164000, + "output": 164000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.27, + "output": 0.41 } }, - "anthropic/claude-3-haiku": { - "id": "anthropic/claude-3-haiku", - "name": "Claude Haiku 3", - "family": "claude-haiku", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-13", - "last_updated": "2024-03-13", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 4096 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude Opus 4 (latest)", - "family": "claude-opus", - "attachment": true, + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1.74, + "output": 3.48, + "cache_read": 0.145 } }, - "anthropic/claude-opus-4-6": { - "id": "anthropic/claude-opus-4-6", - "name": "Claude Opus 4.6 (latest)", - "family": "claude-opus", - "attachment": true, + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "deepseek-ai/DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2025-12-03", + "last_updated": "2025-12-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 164000, + "output": 164000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 0.27, + "output": 0.42 } }, - "anthropic/claude-3.5-haiku": { - "id": "anthropic/claude-3.5-haiku", - "name": "Claude Haiku 3.5 (latest)", - "family": "claude-haiku", - "attachment": true, + "deepseek-ai/DeepSeek-V3": { + "id": "deepseek-ai/DeepSeek-V3", + "name": "deepseek-ai/DeepSeek-V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2024-12-26", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 164000, + "output": 164000 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0.25, + "output": 1 } }, - "anthropic/claude-sonnet-4-5": { - "id": "anthropic/claude-sonnet-4-5", - "name": "Claude Sonnet 4.5 (latest)", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMaxAI/MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-02-15", + "last_updated": "2026-06-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 197000, + "output": 131000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.3, + "output": 1.2 } - }, - "anthropic/claude-3-sonnet": { - "id": "anthropic/claude-3-sonnet", - "name": "Claude Sonnet 3", - "family": "claude-sonnet", + } + } + }, + "umans-ai-coding-plan": { + "id": "umans-ai-coding-plan", + "env": ["UMANS_AI_CODING_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.code.umans.ai/v1", + "name": "Umans AI Coding Plan", + "doc": "https://app.umans.ai/offers/code/docs", + "models": { + "umans-kimi-k2.7": { + "id": "umans-kimi-k2.7", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-04", - "last_updated": "2024-03-04", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 0.3 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "anthropic/claude-3-opus": { - "id": "anthropic/claude-3-opus", - "name": "Claude Opus 3", - "family": "claude-opus", - "attachment": true, - "reasoning": false, + "umans-glm-5.1": { + "id": "umans-glm-5.1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-02-29", - "last_updated": "2024-02-29", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 4096 + "context": 204800, + "output": 131072 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "umans-coder": { + "id": "umans-coder", + "name": "Umans Coder", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 262144, + "output": 262144 }, - "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] - } - } - } - }, - "github-copilot": { - "id": "github-copilot", - "env": ["GITHUB_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.githubcopilot.com", - "name": "GitHub Copilot", - "doc": "https://docs.github.com/en/copilot", - "models": { - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1-Codex-max", - "family": "gpt-codex", + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + }, + "umans-flash": { + "id": "umans-flash", + "name": "Umans Flash", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-12-04", - "last_updated": "2025-12-04", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 128000, - "output": 128000 + "context": 262144, + "output": 262144 }, - "status": "deprecated", "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "claude-opus-4.6": { - "id": "claude-opus-4.6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "umans-glm-5.2": { + "id": "umans-glm-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 144000, - "input": 128000, - "output": 64000 + "context": 405504, + "output": 131072 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "umans-qwen3.6-35b-a3b": { + "id": "umans-qwen3.6-35b-a3b", + "name": "Qwen3.6 35B A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "temperature": false, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 64000 + "context": 262144, + "output": 262144 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 + } + } + } + }, + "io-net": { + "id": "io-net", + "env": ["IOINTELLIGENCE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.intelligence.io.solutions/api/v1", + "name": "IO.NET", + "doc": "https://io.net/docs/guides/intelligence/io-intelligence", + "models": { + "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { + "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", + "name": "Llama 4 Maverick 17B 128E Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-01-15", + "last_updated": "2025-01-15", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 430000, + "output": 4096 + }, + "cost": { + "input": 0.15, + "output": 0.6, + "cache_read": 0.075, + "cache_write": 0.3 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "input": 128000, - "output": 64000 + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.13, + "output": 0.38, + "cache_read": 0.065, + "cache_write": 0.26 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "meta-llama/Llama-3.2-90B-Vision-Instruct": { + "id": "meta-llama/Llama-3.2-90B-Vision-Instruct", + "name": "Llama 3.2 90B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 16000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.35, + "output": 0.4, + "cache_read": 0.175, + "cache_write": 0.7 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5-mini", - "family": "gpt-mini", - "attachment": true, + "moonshotai/Kimi-K2-Thinking": { + "id": "moonshotai/Kimi-K2-Thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-08-13", - "last_updated": "2025-08-13", + "knowledge": "2024-08", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 264000, - "input": 128000, - "output": 64000 + "context": 32768, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.55, + "output": 2.25, + "cache_read": 0.275, + "cache_write": 1.1 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "moonshotai/Kimi-K2-Instruct-0905": { + "id": "moonshotai/Kimi-K2-Instruct-0905", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "knowledge": "2024-08", + "release_date": "2024-09-05", + "last_updated": "2024-09-05", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 64000 + "context": 32768, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.39, + "output": 1.9, + "cache_read": 0.195, + "cache_write": 0.78 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3-Codex", - "family": "gpt-codex", + "Qwen/Qwen2.5-VL-32B-Instruct": { + "id": "Qwen/Qwen2.5-VL-32B-Instruct", + "name": "Qwen 2.5 VL 32B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "temperature": true, + "knowledge": "2024-09", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.05, + "output": 0.22, + "cache_read": 0.025, + "cache_write": 0.1 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, + "Qwen/Qwen3-Next-80B-A3B-Instruct": { + "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", + "name": "Qwen 3 Next 80B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "knowledge": "2024-12", + "release_date": "2025-01-10", + "last_updated": "2025-01-10", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 64000 + "context": 262144, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.8, + "cache_read": 0.05, + "cache_write": 0.2 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen 3 235B Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 264000, - "input": 128000, - "output": 64000 + "context": 262144, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.11, + "output": 0.6, + "cache_read": 0.055, + "cache_write": 0.22 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT-OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.04, + "output": 0.4, + "cache_read": 0.02, + "cache_write": 0.08 } }, - "claude-opus-4.7": { - "id": "claude-opus-4.7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT-OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 144000, - "input": 128000, - "output": 64000 + "context": 64000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.03, + "output": 0.14, + "cache_read": 0.015, + "cache_write": 0.06 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "family": "gpt-codex", + "mistralai/Devstral-Small-2505": { + "id": "mistralai/Devstral-Small-2505", + "name": "Devstral Small 2505", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-05-01", + "last_updated": "2025-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.05, + "output": 0.22, + "cache_read": 0.025, + "cache_write": 0.1 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1-Codex-mini", - "family": "gpt-codex", + "mistralai/Magistral-Small-2506": { + "id": "mistralai/Magistral-Small-2506", + "name": "Magistral Small 2506", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral-small", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 128000, - "output": 128000 + "context": 128000, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 1.5, + "cache_read": 0.25, + "cache_write": 1 } }, - "claude-sonnet-4": { - "id": "claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "mistralai/Mistral-Large-Instruct-2411": { + "id": "mistralai/Mistral-Large-Instruct-2411", + "name": "Mistral Large Instruct 2411", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2024-10", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 216000, - "input": 128000, - "output": 16000 + "context": 128000, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 6, + "cache_read": 1, + "cache_write": 4 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", + "mistralai/Mistral-Nemo-Instruct-2407": { + "id": "mistralai/Mistral-Nemo-Instruct-2407", + "name": "Mistral Nemo Instruct 2407", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-27", - "last_updated": "2025-08-27", + "knowledge": "2024-05", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "input": 128000, - "output": 64000 + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.02, + "output": 0.04, + "cache_read": 0.01, + "cache_write": 0.04 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, - "reasoning": true, + "Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar": { + "id": "Intel/Qwen3-Coder-480B-A35B-Instruct-int4-mixed-ar", + "name": "Qwen 3 Coder 480B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2025-01-15", + "last_updated": "2025-01-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 264000, - "input": 128000, - "output": 64000 + "context": 106000, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.22, + "output": 0.95, + "cache_read": 0.11, + "cache_write": 0.44 } }, - "claude-sonnet-4.5": { - "id": "claude-sonnet-4.5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "zai-org/GLM-4.6": { + "id": "zai-org/GLM-4.6", + "name": "GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2024-10", + "release_date": "2024-11-15", + "last_updated": "2024-11-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 144000, - "input": 128000, - "output": 32000 + "context": 200000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.4, + "output": 1.75, + "cache_read": 0.2, + "cache_write": 0.8 } }, - "claude-opus-41": { - "id": "claude-opus-41", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 80000, - "output": 16000 + "context": 128000, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 8.75, + "cache_read": 1, + "cache_write": 4 } - }, - "claude-opus-4.5": { - "id": "claude-opus-4.5", - "name": "Claude Opus 4.5", - "family": "claude-opus", + } + } + }, + "gmicloud": { + "id": "gmicloud", + "env": ["GMICLOUD_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.gmi-serving.com/v1", + "name": "GMI Cloud", + "doc": "https://docs.gmicloud.ai/inference-engine/api-reference/llm-api-reference", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 160000, - "input": 128000, - "output": 32000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.855, + "output": 3.6, + "cache_read": 0.144 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": false, + "moonshotai/kimi-k2.7-code-highspeed": { + "id": "moonshotai/kimi-k2.7-code-highspeed", + "name": "Kimi K2.7 Code Highspeed", + "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 1.9, + "output": 8, + "cache_read": 0.38 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, - "reasoning": false, + "Qwen/Qwen3.7-Max": { + "id": "Qwen/Qwen3.7-Max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 64000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 7.5, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 1050000, + "input": 922000, "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "claude-haiku-4.5": { - "id": "claude-haiku-4.5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 144000, - "input": 128000, - "output": 32000 + "context": 409600, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 0, - "output": 0 + "input": 4.5, + "output": 22.5, + "cache_read": 0.45 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 64000, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5 } }, - "claude-sonnet-4.6": { - "id": "claude-sonnet-4.6", + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, "knowledge": "2025-08-31", "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 128000, - "output": 32000 + "context": 409600, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1-Codex", - "family": "gpt-codex", - "attachment": false, + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 128000, + "context": 409600, "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5 } - } - } - }, - "mixlayer": { - "id": "mixlayer", - "env": ["MIXLAYER_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://models.mixlayer.ai/v1", - "name": "Mixlayer", - "doc": "https://docs.mixlayer.com", - "models": { - "qwen/qwen3.5-122b-a10b": { - "id": "qwen/qwen3.5-122b-a10b", - "name": "Qwen3.5 122B A10B", - "family": "qwen", + }, + "zai-org/GLM-5.1-FP8": { + "id": "zai-org/GLM-5.1-FP8", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 3.2 + "input": 0.98, + "output": 3.08, + "cache_read": 0.182 } }, - "qwen/qwen3.5-27b": { - "id": "qwen/qwen3.5-27b", - "name": "Qwen3.5 27B", - "family": "qwen", + "zai-org/GLM-5-FP8": { + "id": "zai-org/GLM-5-FP8", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 2.4 + "input": 0.6, + "output": 1.92, + "cache_read": 0.12 } }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", - "family": "qwen", + "zai-org/GLM-5.2-FP8": { + "id": "zai-org/GLM-5.2-FP8", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.979, + "output": 3.08, + "cache_read": 0.182 } }, - "qwen/qwen3.5-9b": { - "id": "qwen/qwen3.5-9b", - "name": "Qwen3.5 9B", - "family": "qwen", + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1048575, + "output": 384000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 0.112, + "output": 0.224, + "cache_read": 0.022 } }, - "qwen/qwen3.5-35b-a3b": { - "id": "qwen/qwen3.5-35b-a3b", - "name": "Qwen3.5 35B A3B", - "family": "qwen", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 0.25, - "output": 1.3 + "input": 1.392, + "output": 2.784, + "cache_read": 0.116 } } } }, - "xiaomi-token-plan-sgp": { - "id": "xiaomi-token-plan-sgp", + "xiaomi-token-plan-cn": { + "id": "xiaomi-token-plan-cn", "env": ["XIAOMI_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://token-plan-sgp.xiaomimimo.com/v1", - "name": "Xiaomi Token Plan (Singapore)", + "api": "https://token-plan-cn.xiaomimimo.com/v1", + "name": "Xiaomi Token Plan (China)", "doc": "https://platform.xiaomimimo.com/#/docs", "models": { - "mimo-v2-tts": { - "id": "mimo-v2-tts", - "name": "MiMo-V2-TTS", + "mimo-v2.5-tts-voiceclone": { + "id": "mimo-v2.5-tts-voiceclone", + "name": "MiMo-V2.5-TTS-VoiceClone", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "family": "mimo", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["audio"] }, "open_weights": true, - "limit": { - "context": 8192, - "output": 16384 - }, - "cost": { - "input": 0, - "output": 0 - } - }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "MiMo-V2-Flash", - "family": "mimo", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 65536 + "limit": { + "context": 8192, + "output": 8192 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0 + "output": 0 } }, - "mimo-v2-pro": { - "id": "mimo-v2-pro", - "name": "MiMo-V2-Pro", + "mimo-v2.5-tts-voicedesign": { + "id": "mimo-v2.5-tts-voicedesign", + "name": "MiMo-V2.5-TTS-VoiceDesign", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "family": "mimo", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 131072 + "context": 8192, + "output": 8192 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0 + "output": 0 } }, "mimo-v2.5": { "id": "mimo-v2.5", "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", "family": "mimo", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -57380,12 +103751,42 @@ "cache_read": 0 } }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo-V2-Omni", + "mimo-v2-tts": { + "id": "mimo-v2-tts", + "name": "MiMo-V2-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "family": "mimo", - "attachment": true, + "attachment": false, + "reasoning": false, + "tool_call": false, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", + "modalities": { + "input": ["text"], + "output": ["audio"] + }, + "open_weights": true, + "limit": { + "context": 8192, + "output": 8192 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "mimo-v2-pro": { + "id": "mimo-v2-pro", + "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -57395,14 +103796,15 @@ "release_date": "2026-03-18", "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, + "context": 1048576, "output": 131072 }, + "status": "deprecated", "cost": { "input": 0, "output": 0, @@ -57412,9 +103814,15 @@ "mimo-v2.5-pro": { "id": "mimo-v2.5-pro", "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -57437,2076 +103845,2230 @@ "output": 0, "cache_read": 0 } - } - } - }, - "zai": { - "id": "zai", - "env": ["ZHIPU_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.z.ai/api/paas/v4", - "name": "Z.AI", - "doc": "https://docs.z.ai/guides/overview/pricing", - "models": { - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM-5V-Turbo", - "family": "glm", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", - "modalities": { - "input": ["text", "image", "video", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24, - "cache_write": 0 - } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "mimo-v2.5-tts": { + "id": "mimo-v2.5-tts", + "name": "MiMo-V2.5-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 0, + "output": 0 } - }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", + } + } + }, + "zeldoc": { + "id": "zeldoc", + "env": ["ZELDOC_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.zeldoc.ai/v1", + "name": "Zeldoc", + "doc": "https://docs.zeldoc.ai", + "models": { + "z-code": { + "id": "z-code", + "name": "Z-Code", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2025-01", + "release_date": "2026-04-15", + "last_updated": "2026-04-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, + "context": 1000000, "output": 131072 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2, - "cache_write": 0 + "input": 0, + "output": 0 } - }, - "glm-4.7-flashx": { - "id": "glm-4.7-flashx", - "name": "GLM-4.7-FlashX", - "family": "glm-flash", + } + } + }, + "scaleway": { + "id": "scaleway", + "env": ["SCALEWAY_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.scaleway.ai/v1", + "name": "Scaleway", + "doc": "https://www.scaleway.com/en/docs/generative-apis/", + "models": { + "qwen3-235b-a22b-instruct-2507": { + "id": "qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Large open Qwen MoE for multilingual reasoning, coding, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-07-01", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 260000, + "output": 16384 }, "cost": { - "input": 0.07, - "output": 0.4, - "cache_read": 0.01, - "cache_write": 0 + "input": 0.75, + "output": 2.25, + "reasoning": 8.4 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "output": 32768 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26, - "cache_write": 0 + "input": 0.2, + "output": 0.8 } }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM-4.5", - "family": "glm", + "qwen3-embedding-8b": { + "id": "qwen3-embedding-8b", + "name": "Qwen3 Embedding 8B", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-25-11", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 32768, + "output": 4096 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 0.1, + "output": 0 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM-4.5-Air", - "family": "glm-air", + "bge-multilingual-gemma2": { + "id": "bge-multilingual-gemma2", + "name": "BGE Multilingual Gemma2", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2024-07-26", + "last_updated": "2025-06-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 8191, + "output": 3072 }, "cost": { - "input": 0.2, - "output": 1.1, - "cache_read": 0.03, - "cache_write": 0 + "input": 0.1, + "output": 0 } }, - "glm-5-turbo": { - "id": "glm-5-turbo", - "name": "GLM-5-Turbo", - "family": "glm", - "attachment": false, + "qwen3.6-35b-a3b": { + "id": "qwen3.6-35b-a3b", + "name": "Qwen3.6 35B A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-04", + "release_date": "2026-05-01", + "last_updated": "2026-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "output": 16384 }, + "status": "beta", "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24, - "cache_write": 0 + "input": 0.25, + "output": 1.5 } }, - "glm-4.5v": { - "id": "glm-4.5v", - "name": "GLM-4.5V", - "family": "glm", + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 64000, + "context": 100000, "output": 16384 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 0.9, + "output": 0.9 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "GLM-4.6", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 1.8, + "output": 5.5 } }, - "glm-4.6v": { - "id": "glm-4.6v", - "name": "GLM-4.6V", - "family": "glm", + "pixtral-12b-2409": { + "id": "pixtral-12b-2409", + "name": "Pixtral 12B 2409", + "description": "Mistral vision-language model for image understanding and multimodal chat", + "family": "pixtral", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "knowledge": "2024-09", + "release_date": "2024-09-25", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 128000, - "output": 32768 + "output": 4096 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.2, + "output": 0.2 } }, - "glm-4.5-flash": { - "id": "glm-4.5-flash", - "name": "GLM-4.5-Flash", - "family": "glm-flash", + "mistral-small-3.2-24b-instruct-2506": { + "id": "mistral-small-3.2-24b-instruct-2506", + "name": "Mistral Small 3.2 24B Instruct (2506)", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-03", + "release_date": "2025-06-20", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.15, + "output": 0.35 } }, - "glm-4.7-flash": { - "id": "glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", - "attachment": false, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT-OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2024-01-01", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.15, + "output": 0.6 } - } - } - }, - "opencode": { - "id": "opencode", - "env": ["OPENCODE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://opencode.ai/zen/v1", - "name": "OpenCode Zen", - "doc": "https://opencode.ai/docs/zen", - "models": { - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "MiniMax M2.7", - "family": "minimax", - "attachment": false, + }, + "gemma-4-26b-a4b-it": { + "id": "gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-04", + "release_date": "2026-04-01", + "last_updated": "2026-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "output": 16384 }, + "status": "beta", "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0.25, + "output": 0.5 } }, - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", - "family": "gpt-codex", + "mistral-medium-3.5-128b": { + "id": "mistral-medium-3.5-128b", + "name": "Mistral Medium 3.5 128B", + "description": "Balanced Mistral model for enterprise assistants, multilingual work, and tools", + "family": "mistral-medium", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1.5, + "output": 7.5 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-04", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 256000, + "output": 16384 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.6, + "output": 3.6 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "whisper-large-v3": { + "id": "whisper-large-v3", + "name": "Whisper Large v3", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", + "family": "whisper", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2023-09-01", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 0, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.08 + "input": 0.003, + "output": 0 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "gemma-3-27b-it": { + "id": "gemma-3-27b-it", + "name": "Gemma-3-27B-IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-12", + "release_date": "2024-12-01", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 40000, + "output": 8192 }, - "status": "deprecated", "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.1 + "input": 0.25, + "output": 0.5 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, - "reasoning": true, + "voxtral-small-24b-2507": { + "id": "voxtral-small-24b-2507", + "name": "Voxtral Small 24B 2507", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "voxtral", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2025-07-01", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 32000, + "output": 16384 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.15, + "output": 0.35 } }, - "glm-4.7-free": { - "id": "glm-4.7-free", - "name": "GLM-4.7 Free", - "family": "glm-free", + "devstral-2-123b-instruct-2512": { + "id": "devstral-2-123b-instruct-2512", + "name": "Devstral 2 123B Instruct (2512)", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2025-12", + "release_date": "2026-01-07", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "output": 16384 }, "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.4, + "output": 2 } - }, - "gemini-3.1-pro": { - "id": "gemini-3.1-pro", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + } + } + }, + "empiriolabs": { + "id": "empiriolabs", + "env": ["EMPIRIOLABS_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.empiriolabs.ai/v1", + "name": "EmpirioLabs AI", + "doc": "https://docs.empiriolabs.ai", + "models": { + "qwen3-5-plus": { + "id": "qwen3-5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1000000, "output": 65536 }, - "provider": { - "npm": "@ai-sdk/google" - }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, + "input": 0.36, + "output": 2.21, "tiers": [ { - "input": 4, - "output": 18, - "cache_read": 0.4, + "input": 1.08, + "output": 6.62, "tier": { "type": "context", - "size": 200000 + "size": 256000 } } ], "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 + "input": 1.08, + "output": 6.62 } } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 393216 + } + ], "tool_call": true, - "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "output": 393216 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.14, + "output": 0.28 } }, - "kimi-k2.5-free": { - "id": "kimi-k2.5-free", - "name": "Kimi K2.5 Free", - "family": "kimi-free", + "kimi-k2-6": { + "id": "kimi-k2-6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-06-12", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 256000, + "output": 16000 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.8939, + "output": 3.7131, + "cache_read": 0.1788 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "qwen3-5-27b": { + "id": "qwen3-5-27b", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 80000 + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 256000, + "output": 64000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.086, + "output": 0.688, + "tiers": [ + { + "input": 0.258, + "output": 2.064, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", + "step-3-7-flash": { + "id": "step-3-7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "input": 256000, + "output": 131072 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", + "qwen3-5-397b-a17b": { + "id": "qwen3-5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 80000 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" - }, - "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 - } - }, - "minimax-m2.5-free": { - "id": "minimax-m2.5-free", - "name": "MiniMax M2.5 Free", - "family": "minimax-free", - "attachment": false, - "reasoning": true, - "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 256000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.172, + "output": 1.032, + "tiers": [ + { + "input": 0.43, + "output": 2.58, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "ring-2.6-1t-free": { - "id": "ring-2.6-1t-free", - "name": "Ring 2.6 1T Free", - "family": "ring-1t-free", - "attachment": false, + "qwen3-5-35b-a3b": { + "id": "qwen3-5-35b-a3b", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 80000 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-05-08", - "last_updated": "2026-05-08", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 66000 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.057, + "output": 0.459, + "tiers": [ + { + "input": 0.229, + "output": 1.835, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.08, + "output": 5.52, + "tiers": [ + { + "input": 2.7, + "output": 13.8, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "deepseek-v4-flash-free": { - "id": "deepseek-v4-flash-free", - "name": "DeepSeek V4 Flash Free", - "family": "deepseek-flash-free", + "glm-5-1": { + "id": "glm-5-1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 38912 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-04-07", + "last_updated": "2026-06-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 202000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.825, + "output": 3.301, + "cache_read": 0.165, + "tiers": [ + { + "input": 1.1, + "output": 3.851, + "cache_read": 0.165, + "tier": { + "type": "context", + "size": 32000 + } + } + ] } }, - "big-pickle": { - "id": "big-pickle", - "name": "Big Pickle", - "family": "big-pickle", - "attachment": false, + "qwen3-7-plus": { + "id": "qwen3-7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 256000 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-10-17", - "last_updated": "2025-10-17", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.4, + "output": 1.6, + "tiers": [ + { + "input": 1.2, + "output": 4.8, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.2, + "output": 4.8 + } } }, - "claude-opus-4-1": { - "id": "claude-opus-4-1", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "kimi-k2-7-code": { + "id": "kimi-k2-7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 256000, + "output": 131072 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.95, + "output": 4 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen3.6", - "attachment": true, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 393216 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 1000000, + "output": 393216 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625 + "input": 1.65, + "output": 3.3 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, + "glm-4-5-flash": { + "id": "glm-4-5-flash", + "name": "GLM 4.5 Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 200000, + "output": 98304 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0, + "output": 0 } }, - "claude-3-5-haiku": { - "id": "claude-3-5-haiku", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", + "qwen3-6-flash": { + "id": "qwen3-6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 64000 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 - }, - "status": "deprecated", - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0.25, + "output": 1.5, + "tiers": [ + { + "input": 1, + "output": 4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1, + "output": 4 + } } }, - "minimax-m2.1": { - "id": "minimax-m2.1", - "name": "MiniMax M2.1", + "minimax-m2-7-highspeed": { + "id": "minimax-m2-7-highspeed", + "name": "MiniMax M2.7 Highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 32768 }, - "status": "deprecated", "cost": { "input": 0.3, "output": 1.2, - "cache_read": 0.1 + "cache_read": 0.03 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "minimax-m2-7": { + "id": "minimax-m2-7", + "name": "MiniMax M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 32768 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.15, + "output": 0.6, + "cache_read": 0.03 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", + "kimi-k2-7-code-highspeed": { + "id": "kimi-k2-7-code-highspeed", + "name": "Kimi K2.7 Code Highspeed", + "description": "Lower-latency Kimi Code variant for interactive edits and coding-agent loops", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 1.9, + "output": 8 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", + "minimax-m3": { + "id": "minimax-m3", + "name": "MiniMax M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 1000000, + "output": 524288 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.225, + "output": 0.9, + "cache_read": 0.045, + "tiers": [ + { + "input": 0.45, + "output": 1.8, + "cache_read": 0.045, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.45, + "output": 1.8, + "cache_read": 0.045 + } } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex Mini", - "family": "gpt-codex", + "mimo-v2-5": { + "id": "mimo-v2-5", + "name": "MiMo V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, + "context": 1000000, "output": 128000 }, - "provider": { - "npm": "@ai-sdk/openai" - }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.7, + "output": 1.4, + "cache_read": 0.014 } }, - "claude-sonnet-4": { - "id": "claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "qwen3-6-plus": { + "id": "qwen3-6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, + "input": 0.5, + "output": 3, "tiers": [ { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, + "input": 2, + "output": 6, "tier": { "type": "context", - "size": 200000 + "size": 256000 } } ], "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 + "input": 2, + "output": 6 } } }, - "gemini-3-flash": { - "id": "gemini-3-flash", - "name": "Gemini 3 Flash", - "family": "gemini-flash", - "attachment": true, + "step-3-5-flash": { + "id": "step-3-5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", - "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1048576, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/google" - }, - "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 - } - }, - "trinity-large-preview-free": { - "id": "trinity-large-preview-free", - "name": "Trinity Large Preview", - "family": "trinity", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-28", - "last_updated": "2026-01-28", + "release_date": "2026-01-29", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 256000, + "input": 256000, "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "qwen3-6-27b": { + "id": "qwen3-6-27b", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 80000 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "output": 64000 }, "cost": { - "input": 1.07, - "output": 8.5, - "cache_read": 0.107 + "input": 0.412564, + "output": 2.475384 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", - "attachment": true, + "glm-5-2": { + "id": "glm-5-2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 1000000, + "output": 131072 }, "cost": { - "input": 30, - "output": 180, - "cache_read": 30 + "input": 1.4, + "output": 4.4 } }, - "glm-5-free": { - "id": "glm-5-free", - "name": "GLM-5 Free", - "family": "glm-free", + "glm-4-7-flash": { + "id": "glm-4-7-flash", + "name": "GLM 4.7 Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 200000, "output": 131072 }, - "status": "deprecated", "cost": { "input": 0, - "output": 0, - "cache_read": 0 + "output": 0 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "step-3-5-flash-2603": { + "id": "step-3-5-flash-2603", + "name": "Step 3.5 Flash 2603", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 256000, + "input": 256000, + "output": 131072 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "minimax-m2.1-free": { - "id": "minimax-m2.1-free", - "name": "MiniMax M2.1 Free", - "family": "minimax-free", - "attachment": false, + "qwen3-5-4b": { + "id": "qwen3-5-4b", + "name": "Qwen3.5 4B", + "description": "Qwen3.5 4B is a low-cost multimodal reasoning model with 256K context, image and video input, function tools, and structured output.", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2026-03-02", + "last_updated": "2026-03-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 - }, - "status": "deprecated", - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.04, + "output": 0.07, + "cache_read": 0.02 } }, - "qwen3.6-plus-free": { - "id": "qwen3.6-plus-free", - "name": "Qwen3.6 Plus Free", - "family": "qwen-free", - "attachment": false, + "gemma-4-26b-a4b": { + "id": "gemma-4-26b-a4b", + "name": "Gemma 4 26B-A4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "release_date": "2026-04-02", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 64000 + "context": 262144, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.05, + "output": 0.29, + "cache_read": 0.025 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "GLM-4.6", - "family": "glm", - "attachment": false, + "qwen3-5-9b": { + "id": "qwen3-5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.1 + "input": 0.09, + "output": 0.13, + "cache_read": 0.045 } }, - "ling-2.6-flash-free": { - "id": "ling-2.6-flash-free", - "name": "Ling 2.6 Flash Free", - "family": "ling-flash-free", + "qwen3-7-max": { + "id": "qwen3-7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 64000 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-05-21", + "last_updated": "2026-06-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262100, - "output": 32800 + "context": 1000000, + "output": 65536 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 7.5 } }, - "gemini-3-pro": { - "id": "gemini-3-pro", - "name": "Gemini 3 Pro", - "family": "gemini-pro", - "attachment": true, + "qwen3-6-max-preview": { + "id": "qwen3-6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 393216 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 256000, "output": 65536 }, - "status": "deprecated", - "provider": { - "npm": "@ai-sdk/google" - }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, + "input": 1.31, + "output": 7.88, "tiers": [ { - "input": 4, - "output": 18, - "cache_read": 0.4, + "input": 1.97, + "output": 11.82, "tier": { "type": "context", - "size": 200000 + "size": 128000 } } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + ] } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", - "attachment": true, + "mimo-v2-5-pro": { + "id": "mimo-v2-5-pro", + "name": "MiMo V2.5 Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 2.175, + "output": 4.35, + "cache_read": 0.018 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "GPT-5 Codex", - "family": "gpt-codex", + "qwen3-5-122b-a10b": { + "id": "qwen3-5-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1, + "max": 80000 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 256000, + "output": 64000 }, "cost": { - "input": 1.07, - "output": 8.5, - "cache_read": 0.107 + "input": 0.115, + "output": 0.917, + "tiers": [ + { + "input": 0.287, + "output": 2.294, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "grok-code": { - "id": "grok-code", - "name": "Grok Code Fast 1", - "family": "grok", + "fugu-ultra": { + "id": "fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "fugu", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-20", - "last_updated": "2025-08-20", + "structured_output": true, + "temperature": false, + "release_date": "2026-06-15", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 1000000, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 7.5, + "output": 45, + "cache_read": 1.5, + "tiers": [ + { + "input": 15, + "output": 67.5, + "cache_read": 1.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 15, + "output": 67.5, + "cache_read": 1.5 + } } }, - "mimo-v2-flash-free": { - "id": "mimo-v2-flash-free", - "name": "MiMo V2 Flash Free", - "family": "mimo-flash-free", - "attachment": false, + "muse-spark-1-1": { + "id": "muse-spark-1-1", + "name": "Muse Spark 1.1", + "description": "Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.", + "family": "muse", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "release_date": "2026-04-08", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 1.25, + "output": 4.25, + "cache_read": 1 } - }, - "gpt-5.3-codex-spark": { - "id": "gpt-5.3-codex-spark", - "name": "GPT-5.3 Codex Spark", - "family": "gpt-codex-spark", + } + } + }, + "ovhcloud": { + "id": "ovhcloud", + "env": ["OVHCLOUD_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://oai.endpoints.kepler.ai.cloud.ovh.net/v1", + "name": "OVHcloud AI Endpoints", + "doc": "https://www.ovhcloud.com/en/public-cloud/ai-endpoints/catalog//", + "models": { + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder-30B-A3B-Instruct", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "temperature": true, + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 128000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.07, + "output": 0.26 } }, - "hy3-preview-free": { - "id": "hy3-preview-free", - "name": "Hy3 preview Free", - "family": "hy3-free", + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3-32B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "release_date": "2025-07-16", + "last_updated": "2025-07-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 + "context": 32768, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.09, + "output": 0.25 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", + "qwen3guard-gen-8b": { + "id": "qwen3guard-gen-8b", + "name": "Qwen3Guard-Gen-8B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "context": 32768, + "output": 16384 } }, - "kimi-k2": { - "id": "kimi-k2", - "name": "Kimi K2", - "family": "kimi", + "qwen3guard-gen-0.6b": { + "id": "qwen3guard-gen-0.6b", + "name": "Qwen3Guard-Gen-0.6B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 - }, - "status": "deprecated", - "cost": { - "input": 0.4, - "output": 2.5, - "cache_read": 0.4 - } - }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "context": 32768, + "output": 16384 } }, - "qwen3-coder": { - "id": "qwen3-coder", - "name": "Qwen3 Coder", - "family": "qwen", + "meta-llama-3_3-70b-instruct": { + "id": "meta-llama-3_3-70b-instruct", + "name": "Meta-Llama-3_3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2025-04-01", + "last_updated": "2025-04-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0.45, - "output": 1.8 + "input": 0.74, + "output": 0.74 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", + "mistral-small-3.2-24b-instruct-2506": { + "id": "mistral-small-3.2-24b-instruct-2506", + "name": "Mistral-Small-3.2-24B-Instruct-2506", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "release_date": "2025-07-16", + "last_updated": "2025-07-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 131072, + "output": 131072 }, "cost": { - "input": 1.07, - "output": 8.5, - "cache_read": 0.107 + "input": 0.1, + "output": 0.31 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", - "family": "qwen3.5", + "qwen2.5-vl-72b-instruct": { + "id": "qwen2.5-vl-72b-instruct", + "name": "Qwen2.5-VL-72B-Instruct", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "release_date": "2025-03-31", + "last_updated": "2025-03-31", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 65536 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 1.2, - "cache_read": 0.02, - "cache_write": 0.25 + "input": 1.01, + "output": 1.01 } }, - "mimo-v2-pro-free": { - "id": "mimo-v2-pro-free", - "name": "MiMo V2 Pro Free", - "family": "mimo-pro-free", - "attachment": true, + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 64000 + "context": 131072, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.09, + "output": 0.47 } }, - "nemotron-3-super-free": { - "id": "nemotron-3-super-free", - "name": "Nemotron 3 Super Free", - "family": "nemotron-free", + "mistral-7b-instruct-v0.3": { + "id": "mistral-7b-instruct-v0.3", + "name": "Mistral-7B-Instruct-v0.3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2026-02", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "release_date": "2025-04-01", + "last_updated": "2025-04-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 128000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.11, + "output": 0.11 } }, - "gpt-5.5-pro": { - "id": "gpt-5.5-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "mistral-nemo-instruct-2407": { + "id": "mistral-nemo-instruct-2407", + "name": "Mistral-Nemo-Instruct-2407", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "structured_output": true, + "temperature": true, + "release_date": "2024-11-20", + "last_updated": "2024-11-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 65536, + "output": 65536 }, "cost": { - "input": 30, - "output": 180, - "cache_read": 30 + "input": 0.14, + "output": 0.14 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "qwen3.6-27b": { + "id": "qwen3.6-27b", + "name": "Qwen3.6-27B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, @@ -59514,480 +106076,492 @@ "context": 262144, "output": 262144 }, - "status": "deprecated", "cost": { - "input": 0.4, - "output": 2.5, - "cache_read": 0.4 + "input": 0.47, + "output": 3.19 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "qwen3.5-9b": { + "id": "qwen3.5-9b", + "name": "Qwen3.5-9B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.07, - "output": 8.5, - "cache_read": 0.107 + "input": 0.12, + "output": 0.18 } }, - "mimo-v2-omni-free": { - "id": "mimo-v2-omni-free", - "name": "MiMo V2 Omni Free", - "family": "mimo-omni-free", + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5-397B-A17B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-05-18", + "last_updated": "2026-05-18", "modalities": { - "input": ["text", "image", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 64000 + "output": 262144 }, - "status": "deprecated", "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 0.71, + "output": 4.25 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 131072, + "output": 131072 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.05, + "output": 0.18 } - }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + } + } + }, + "friendli": { + "id": "friendli", + "env": ["FRIENDLI_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.friendli.ai/serverless/v1", + "name": "Friendli", + "doc": "https://friendli.ai/docs/guides/serverless_endpoints/introduction", + "models": { + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai" + "context": 262144, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.14, + "output": 0.4 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Large open Qwen MoE for multilingual reasoning, coding, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "knowledge": "2025-04", + "release_date": "2025-07-29", + "last_updated": "2026-01-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/anthropic" + "context": 262144, + "output": 262144 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.2, + "output": 0.8 } - } - } - }, - "stepfun": { - "id": "stepfun", - "env": ["STEPFUN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.stepfun.com/v1", - "name": "StepFun", - "doc": "https://platform.stepfun.com/docs/zh/overview/concept", - "models": { - "step-3.5-flash-2603": { - "id": "step-3.5-flash-2603", - "name": "Step 3.5 Flash 2603", + }, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 256000 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.02 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "step-1-32k": { - "id": "step-1-32k", - "name": "Step 1 (32K)", + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-01-01", - "last_updated": "2026-02-13", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 202752, + "output": 202752 }, "cost": { - "input": 2.05, - "output": 9.59, - "cache_read": 0.41 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "step-3.5-flash": { - "id": "step-3.5-flash", - "name": "Step 3.5 Flash", + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-29", - "last_updated": "2026-02-13", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 256000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.096, - "output": 0.288, - "cache_read": 0.019 + "input": 0.5, + "output": 1.5, + "cache_read": 0.25 } }, - "step-2-16k": { - "id": "step-2-16k", - "name": "Step 2 (16K)", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-01-01", - "last_updated": "2026-02-13", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, - "input": 16384, - "output": 8192 + "context": 196608, + "output": 196608 }, "cost": { - "input": 5.21, - "output": 16.44, - "cache_read": 1.04 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } } } }, - "nebius": { - "id": "nebius", - "env": ["NEBIUS_API_KEY"], + "tencent-tokenhub": { + "id": "tencent-tokenhub", + "env": ["TENCENT_TOKENHUB_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.tokenfactory.nebius.com/v1", - "name": "Nebius Token Factory", - "doc": "https://docs.tokenfactory.nebius.com/", + "api": "https://tokenhub.tencentmaas.com/v1", + "name": "Tencent TokenHub", + "doc": "https://cloud.tencent.com/document/product/1823/130050", "models": { - "NousResearch/Hermes-4-70B": { - "id": "NousResearch/Hermes-4-70B", - "name": "Hermes-4-70B", + "hy3": { + "id": "hy3", + "name": "Hy3", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2026-01-30", - "last_updated": "2026-02-04", + "release_date": "2026-07-06", + "last_updated": "2026-07-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.13, - "output": 0.4, - "reasoning": 0.4, - "cache_read": 0.013, - "cache_write": 0.16 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "NousResearch/Hermes-4-405B": { - "id": "NousResearch/Hermes-4-405B", - "name": "Hermes-4-405B", + "hy3-preview": { + "id": "hy3-preview", + "name": "Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2026-01-30", - "last_updated": "2026-02-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 256000, + "output": 64000 }, "cost": { - "input": 1, - "output": 3, - "reasoning": 3, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "Qwen/Qwen2.5-VL-72B-Instruct": { - "id": "Qwen/Qwen2.5-VL-72B-Instruct", - "name": "Qwen2.5-VL-72B-Instruct", - "attachment": true, + } + } + }, + "wandb": { + "id": "wandb", + "env": ["WANDB_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.inference.wandb.ai/v1", + "name": "Weights & Biases", + "doc": "https://docs.wandb.ai/guides/integrations/inference/", + "models": { + "ibm-granite/granite-4.1-8b": { + "id": "ibm-granite/granite-4.1-8b", + "name": "Granite 4.1 8B", + "description": "Granite 4.1 8B is a long-context instruct model capable of enhanced tool calling, instruction following, and chat capabilities.", + "family": "granite", + "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-20", - "last_updated": "2026-02-04", + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025, - "cache_write": 0.31 + "input": 0.05, + "output": 0.1, + "cache_read": 0.05 } }, - "Qwen/Qwen3.5-397B-A17B": { - "id": "Qwen/Qwen3.5-397B-A17B", - "name": "Qwen3.5-397B-A17B", + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama 3.3 70B", + "description": "Multilingual model excelling in conversational tasks, detailed instruction-following, and coding.", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-15", - "last_updated": "2026-05-07", + "knowledge": "2023-12", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "input": 250000, - "output": 8192 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3.6, - "cache_read": 0.06, - "cache_write": 0.75 + "input": 0.71, + "output": 0.71, + "cache_read": 0.71 } }, - "Qwen/Qwen3-Embedding-8B": { - "id": "Qwen/Qwen3-Embedding-8B", - "name": "Qwen3-Embedding-8B", - "family": "text-embedding", + "meta-llama/Llama-3.1-70B-Instruct": { + "id": "meta-llama/Llama-3.1-70B-Instruct", + "name": "Llama 3.1 70B", + "description": "Efficient conversational model optimized for responsive multilingual chatbot interactions.", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": false, - "knowledge": "2025-10", - "release_date": "2026-01-10", - "last_updated": "2026-02-04", + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "input": 32768, - "output": 0 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.01, - "output": 0 + "input": 0.8, + "output": 0.8, + "cache_read": 0.8 } }, - "Qwen/Qwen3-30B-A3B-Instruct-2507": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen3-30B-A3B-Instruct-2507", + "meta-llama/Llama-3.1-8B-Instruct": { + "id": "meta-llama/Llama-3.1-8B-Instruct", + "name": "Llama 3.1 8B", + "description": "Efficient conversational model optimized for responsive multilingual chatbot interactions.", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-01-28", - "last_updated": "2026-02-04", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] @@ -59995,1616 +106569,1948 @@ "open_weights": true, "limit": { "context": 128000, - "input": 120000, - "output": 8192 + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01, - "cache_write": 0.125 + "input": 0.22, + "output": 0.22, + "cache_read": 0.22 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi K2.6 is a multimodal Mixture-of-Experts language model featuring 32 billion activated parameters and a total of 1 trillion parameters.", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-25", - "last_updated": "2025-10-04", + "knowledge": "2025-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 262144, - "output": 8192 + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "Qwen/Qwen3-32B": { - "id": "Qwen/Qwen3-32B", - "name": "Qwen3-32B", - "attachment": false, - "reasoning": false, + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Kimi K2.5 is a multimodal Mixture-of-Experts language model featuring 32 billion activated parameters and a total of 1 trillion parameters.", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-01-28", - "last_updated": "2026-02-04", + "knowledge": "2025-01", + "release_date": "2026-02-02", + "last_updated": "2026-02-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01, - "cache_write": 0.125 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507-fast": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507-fast", - "name": "Qwen3-235B-A22B-Thinking-2507-fast", - "attachment": false, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Kimi K2.7 Code is a 1T-parameter MoE model with 32B active parameters purpose-built for long-horizon agentic coding and software engineering.", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-25", - "last_updated": "2026-05-07", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.5, - "output": 2, - "cache_read": 0.05, - "cache_write": 0.625 + "input": 0.94, + "output": 4, + "cache_read": 0.19 } }, - "Qwen/Qwen3.5-397B-A17B-fast": { - "id": "Qwen/Qwen3.5-397B-A17B-fast", - "name": "Qwen3.5-397B-A17B-fast", - "attachment": false, + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B", + "description": "Gemma 4 31B Dense is designed for advanced reasoning, agentic workflows, and longer context and is natively trained on 140+ languages.", + "family": "gemma", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-15", - "last_updated": "2026-05-07", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.6, - "output": 3.6, - "cache_read": 0.06, - "cache_write": 0.75 + "input": 0.12, + "output": 0.35, + "cache_read": 0.09 } }, - "Qwen/Qwen3-Next-80B-A3B-Thinking-fast": { - "id": "Qwen/Qwen3-Next-80B-A3B-Thinking-fast", - "name": "Qwen3-Next-80B-A3B-Thinking-fast", + "microsoft/Phi-4-mini-instruct": { + "id": "microsoft/Phi-4-mini-instruct", + "name": "Phi 4 Mini 3.8B", + "description": "Compact, efficient model ideal for fast responses in resource-constrained environments.", + "family": "phi", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-25", - "last_updated": "2026-05-07", + "knowledge": "2023-10", + "release_date": "2025-02-01", + "last_updated": "2025-02-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 128000, + "output": 128000 }, + "status": "deprecated", "cost": { - "input": 0.15, - "output": 1.2, - "cache_read": 0.015, - "cache_write": 0.1875 + "input": 0.08, + "output": 0.35, + "cache_read": 0.08 } }, - "Qwen/Qwen3-Next-80B-A3B-Thinking": { - "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "name": "Qwen3-Next-80B-A3B-Thinking", - "attachment": false, + "Qwen/Qwen3.6-35B-A3B": { + "id": "Qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B A3B", + "description": "Qwen3.6-35B-A3B is an MoE multimodal model with 262K context optimized for agentic coding workflows.", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-01-28", - "last_updated": "2026-02-04", + "release_date": "2026-04-15", + "last_updated": "2026-04-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.15, - "output": 1.2, - "reasoning": 1.2, - "cache_read": 0.015, - "cache_write": 0.18 + "input": 0.25, + "output": 1.25, + "cache_read": 0.25 } }, - "PrimeIntellect/INTELLECT-3": { - "id": "PrimeIntellect/INTELLECT-3", - "name": "INTELLECT-3", - "attachment": false, - "reasoning": false, + "Qwen/Qwen3.6-27B": { + "id": "Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen3.6-27B is a 27B dense multimodal model with 262K context built for flagship-level agentic coding.", + "family": "qwen3.6", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-10", - "release_date": "2026-01-25", - "last_updated": "2026-02-04", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 1.1, - "cache_read": 0.02, - "cache_write": 0.25 + "input": 0.6, + "output": 3.6, + "cache_read": 0.12 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", + "Qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen3 235B A22B Thinking-2507", + "description": "High-performance Mixture-of-Experts model optimized for structured reasoning, math, and long-form generation.", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-03-01", - "last_updated": "2026-03-10", + "knowledge": "2025-04", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "input": 200000, - "output": 16384 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.1, - "cache_write": 1 + "input": 0.1, + "output": 0.1, + "cache_read": 0.1 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama-3.3-70B-Instruct", + "Qwen/Qwen3-Coder-480B-A35B-Instruct": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", + "name": "Qwen3 Coder 480B A35B", + "description": "Mixture-of-Experts model optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning.", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-12-05", - "last_updated": "2026-02-04", + "knowledge": "2025-04", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.13, - "output": 0.4, - "cache_read": 0.013, - "cache_write": 0.16 + "input": 1, + "output": 1.5, + "cache_read": 1 } }, - "meta-llama/Meta-Llama-3.1-8B-Instruct": { - "id": "meta-llama/Meta-Llama-3.1-8B-Instruct", - "name": "Meta-Llama-3.1-8B-Instruct", - "attachment": false, - "reasoning": false, + "Qwen/Qwen3.5-27B": { + "id": "Qwen/Qwen3.5-27B", + "name": "Qwen3.5-27B", + "description": "Qwen3.5-27B is a dense model from the Qwen3.5 family built for high performance across a large range of benchmarks.", + "family": "qwen3.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-07-23", - "last_updated": "2026-02-04", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 4096 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.02, - "output": 0.06, - "cache_read": 0.002, - "cache_write": 0.025 + "input": 0.39, + "output": 3.12, + "cache_read": 0.08 } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "Nemotron-3-Super-120B-A12B", + "Qwen/Qwen3-30B-A3B-Instruct-2507": { + "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507", + "description": "Qwen3-30B-A3B-Instruct-2507 is a 30.5B MoE instruction-tuned model with enhanced reasoning, coding, and long-context understanding.", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2026-02", - "release_date": "2026-03-11", - "last_updated": "2026-03-12", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.1, + "output": 0.3, + "cache_read": 0.1 } }, - "nvidia/Llama-3_1-Nemotron-Ultra-253B-v1": { - "id": "nvidia/Llama-3_1-Nemotron-Ultra-253B-v1", - "name": "Llama-3.1-Nemotron-Ultra-253B-v1", + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B-2507", + "description": "Efficient multilingual, Mixture-of-Experts, instruction-tuned model, optimized for logical reasoning.", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-15", - "last_updated": "2026-02-04", + "knowledge": "2025-04", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 120000, - "output": 4096 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.6, - "output": 1.8, - "cache_read": 0.06, - "cache_write": 0.75 + "input": 0.1, + "output": 0.1, + "cache_read": 0.1 } }, - "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B": { - "id": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B", - "name": "Nemotron-3-Nano-30B-A3B", + "Qwen/Qwen3.5-35B-A3B": { + "id": "Qwen/Qwen3.5-35B-A3B", + "name": "Qwen3.5-35B-A3B", + "description": "Qwen3.5-35B-A3B is an open-weights multimodal MoE model built for efficient, high-throughput inference across chat, reasoning, and agentic tasks.", + "family": "qwen3.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.25, + "output": 1.25, + "cache_read": 0.25 + } + }, + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Efficient Mixture-of-Experts model designed for high-reasoning, agentic and general-purpose use cases.", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-08-10", - "last_updated": "2026-02-04", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "input": 30000, - "output": 4096 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.06, - "output": 0.24, - "cache_read": 0.006, - "cache_write": 0.075 + "input": 0.04, + "output": 0.14, + "cache_read": 0.04 } }, - "nvidia/Nemotron-3-Nano-Omni": { - "id": "nvidia/Nemotron-3-Nano-Omni", - "name": "Nemotron-3-Nano-Omni", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "gpt-oss-20b", + "description": "Lower latency Mixture-of-Experts model trained on OpenAI's Harmony response format with reasoning capabilities.", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2026-05-07", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "input": 60000, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.06, - "output": 0.24, - "cache_read": 0.006, - "cache_write": 0.075 + "input": 0.03, + "output": 0.13, + "cache_read": 0.03 } }, - "deepseek-ai/DeepSeek-V3.2-fast": { - "id": "deepseek-ai/DeepSeek-V3.2-fast", - "name": "DeepSeek-V3.2-fast", + "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B": { + "id": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B", + "name": "Nemotron 3 Ultra", + "description": "Nemotron 3 Ultra is a powerful MoE model designed for long-running agents across coding, deep research, and enterprise automation.", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2026-05-07", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.04, - "cache_write": 0.5 + "input": 0.75, + "output": 2.75, + "cache_read": 0.15 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek-V3.2", + "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8": { + "id": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8", + "name": "Nemotron 3 Super", + "description": "Nemotron 3 is a LatentMoE model designed to deliver strong agentic, reasoning, and conversational capabilities.", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2026-01-20", - "last_updated": "2026-02-04", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163000, - "input": 160000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.3, - "output": 0.45, - "reasoning": 0.45, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 0.2, + "output": 0.8, + "cache_read": 0.2 } }, - "openai/gpt-oss-120b-fast": { - "id": "openai/gpt-oss-120b-fast", - "name": "gpt-oss-120b-fast", + "JetBrains/Mellum2-12B-A2.5B-Instruct": { + "id": "JetBrains/Mellum2-12B-A2.5B-Instruct", + "name": "Mellum2 12B A2.5B", + "description": "Mellum2-12B-A2.5B-Instruct is a fast MoE model with 131K context built for coding, tool use, and low-latency AI workflows.", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-06-10", - "last_updated": "2026-05-07", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.5, - "cache_read": 0.01, - "cache_write": 0.125 + "input": 0.05, + "output": 0.1, + "cache_read": 0.05 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "gpt-oss-120b", + "OpenPipe/Qwen3-14B-Instruct": { + "id": "OpenPipe/Qwen3-14B-Instruct", + "name": "Qwen3 14B Instruct", + "description": "An efficient multilingual, dense, instruction-tuned model, optimized by OpenPipe for building agents with finetuning.", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2026-01-10", - "last_updated": "2026-02-04", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "input": 124000, - "output": 8192 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6, - "reasoning": 0.6, - "cache_read": 0.015, - "cache_write": 0.18 + "input": 0.05, + "output": 0.22, + "cache_read": 0.05 } }, - "google/gemma-2-2b-it": { - "id": "google/gemma-2-2b-it", - "name": "Gemma-2-2b-it", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM 5.2", + "description": "GLM-5.2 is a Mixture-of-Experts language model featuring 40 billion activated parameters and a total of 744 billion parameters.", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-07-31", - "last_updated": "2026-02-04", + "release_date": "2026-06-16", + "last_updated": "2026-06-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "input": 8000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.02, - "output": 0.06, - "cache_read": 0.002, - "cache_write": 0.025 + "input": 1.39, + "output": 4.4, + "cache_read": 0.26 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Gemma-3-27b-it", - "attachment": true, - "reasoning": false, + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM 5.1", + "description": "Powerful MoE model for long-horizon agentic engineering and advanced reasoning.", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-10", - "release_date": "2026-01-20", - "last_updated": "2026-02-04", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 110000, - "input": 100000, - "output": 8192 + "context": 202752, + "output": 202752 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01, - "cache_write": 0.125 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "moonshotai/Kimi-K2.5-fast": { - "id": "moonshotai/Kimi-K2.5-fast", - "name": "Kimi-K2.5-fast", - "family": "kimi", - "attachment": true, - "reasoning": true, + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "DeepSeek V3.1", + "description": "A large hybrid model that supports both thinking and non-thinking modes via prompt templates.", + "family": "deepseek", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-15", - "last_updated": "2026-02-04", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 8192 + "context": 161000, + "output": 161000 }, "cost": { - "input": 0.5, - "output": 2.5, - "cache_read": 0.05, - "cache_write": 0.625 + "input": 0.55, + "output": 1.65, + "cache_read": 0.55 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi-K2.5", - "family": "kimi", - "attachment": true, + "deepseek-ai/DeepSeek-V4-Flash": { + "id": "deepseek-ai/DeepSeek-V4-Flash", + "name": "DeepSeek V4 Flash", + "description": "DeepSeek V4-Flash is an MoE model with 1M context length great for coding, reasoning, and agentic workloads.", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-12-15", - "last_updated": "2026-02-04", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "input": 256000, - "output": 8192 + "context": 1048576, + "output": 1048576 }, "cost": { - "input": 0.5, - "output": 2.5, - "reasoning": 2.5, - "cache_read": 0.05, - "cache_write": 0.625 + "input": 0.14, + "output": 0.28, + "cache_read": 0.07 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "DeepSeek V4-Pro is a 1.6T-parameter MoE model with 49B active parameters excelling at advanced reasoning, coding, and complex agentic workloads.", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2026-05-07", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "input": 190000, - "output": 8192 + "context": 1048576, + "output": 1048576 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 1.74, + "output": 3.48, + "cache_read": 0.14 } }, - "MiniMaxAI/MiniMax-M2.5-fast": { - "id": "MiniMaxAI/MiniMax-M2.5-fast", - "name": "MiniMax-M2.5-fast", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax M2.5", + "description": "MoE model with a highly sparse architecture designed for high-throughput and low latency with strong coding capabilities.", + "family": "minimax-m2.5", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2026-05-07", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "input": 7000, - "output": 8192 + "context": 196608, + "output": 196608 }, "cost": { "input": 0.3, "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "cache_read": 0.3 } - }, - "deepseek-ai/DeepSeek-V4-Pro": { - "id": "deepseek-ai/DeepSeek-V4-Pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + } + } + }, + "kuae-cloud-coding-plan": { + "id": "kuae-cloud-coding-plan", + "env": ["KUAE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://coding-plan-endpoint.kuaecloud.net/v1", + "name": "KUAE Cloud Coding Plan", + "doc": "https://docs.mthreads.com/kuaecloud/kuaecloud-doc-online/coding_plan/", + "models": { + "GLM-4.7": { + "id": "GLM-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 3.5, - "cache_read": 0.15 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } } } }, - "poe": { - "id": "poe", - "env": ["POE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.poe.com/v1", - "name": "Poe", - "doc": "https://creator.poe.com/docs/external-applications/openai-compatible-api", + "gitlab": { + "id": "gitlab", + "env": ["GITLAB_TOKEN"], + "npm": "gitlab-ai-provider", + "name": "GitLab Duo", + "doc": "https://docs.gitlab.com/user/duo_agent_platform/", "models": { - "topazlabs-co/topazlabs": { - "id": "topazlabs-co/topazlabs", - "name": "TopazLabs", - "family": "topazlabs", + "duo-chat-opus-4-5": { + "id": "duo-chat-opus-4-5", + "name": "Agentic Chat (Claude Opus 4.5)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2026-01-08", + "last_updated": "2026-01-08", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 204, - "output": 0 + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "novita/kimi-k2.5": { - "id": "novita/kimi-k2.5", - "name": "Kimi-K2.5", + "duo-chat-opus-4-8": { + "id": "duo-chat-opus-4-8", + "name": "Agentic Chat (Claude Opus 4.8)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 262144 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "novita/glm-4.7": { - "id": "novita/glm-4.7", - "name": "glm-4.7", + "duo-chat-opus-4-7": { + "id": "duo-chat-opus-4-7", + "name": "Agentic Chat (Claude Opus 4.7)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 131072 + "context": 1000000, + "output": 64000 }, - "status": "deprecated" + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 + } }, - "novita/glm-5": { - "id": "novita/glm-5", - "name": "GLM-5", + "duo-chat-gpt-5-2-codex": { + "id": "duo-chat-gpt-5-2-codex", + "name": "Agentic Chat (GPT-5.2 Codex)", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0, + "output": 0 } }, - "novita/minimax-m2.1": { - "id": "novita/minimax-m2.1", - "name": "minimax-m2.1", + "duo-chat-fable-5": { + "id": "duo-chat-fable-5", + "name": "Agentic Chat (Claude Fable 5)", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-12-26", - "last_updated": "2025-12-26", + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 131072 + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "novita/glm-4.6": { - "id": "novita/glm-4.6", - "name": "GLM-4.6", - "family": "glm", + "duo-chat-gpt-5-5": { + "id": "duo-chat-gpt-5-5", + "name": "Agentic Chat (GPT-5.5)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-08-31", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 0, "output": 0 } }, - "novita/kimi-k2.6": { - "id": "novita/kimi-k2.6", - "name": "Kimi-K2.6", + "duo-chat-opus-4-6": { + "id": "duo-chat-opus-4-6", + "name": "Agentic Chat (Claude Opus 4.6)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-20", - "last_updated": "2026-05-02", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "input": 262144, - "output": 262144 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.96, - "output": 4.04, - "cache_read": 0.16 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "novita/glm-4.6v": { - "id": "novita/glm-4.6v", - "name": "glm-4.6v", + "duo-chat-gpt-5-4": { + "id": "duo-chat-gpt-5-4", + "name": "Agentic Chat (GPT-5.4)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 32768 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 } }, - "novita/deepseek-v3.2": { - "id": "novita/deepseek-v3.2", - "name": "DeepSeek-V3.2", - "attachment": true, + "duo-chat-gpt-5-codex": { + "id": "duo-chat-gpt-5-codex", + "name": "Agentic Chat (GPT-5 Codex)", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 0 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 0.4, - "cache_read": 0.13 + "input": 0, + "output": 0 } }, - "novita/glm-4.7-flash": { - "id": "novita/glm-4.7-flash", - "name": "glm-4.7-flash", + "duo-chat-gpt-5-4-nano": { + "id": "duo-chat-gpt-5-4-nano", + "name": "Agentic Chat (GPT-5.4 Nano)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 65500 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0, + "output": 0 } }, - "novita/glm-4.7-n": { - "id": "novita/glm-4.7-n", - "name": "glm-4.7-n", + "duo-chat-sonnet-4-6": { + "id": "duo-chat-sonnet-4-6", + "name": "Agentic Chat (Claude Sonnet 4.6)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 131072 + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "novita/kimi-k2-thinking": { - "id": "novita/kimi-k2-thinking", - "name": "kimi-k2-thinking", - "family": "kimi", + "duo-chat-gpt-5-mini": { + "id": "duo-chat-gpt-5-mini", + "name": "Agentic Chat (GPT-5 Mini)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-11-07", - "last_updated": "2025-11-07", + "knowledge": "2024-05-30", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0, "output": 0 } }, - "fireworks-ai/kimi-k2.5-fw": { - "id": "fireworks-ai/kimi-k2.5-fw", - "name": "Kimi-K2.5-FW", + "duo-chat-sonnet-5": { + "id": "duo-chat-sonnet-5", + "name": "Agentic Chat (Claude Sonnet 5)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "input": 245760, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { "input": 0, - "output": 0 + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "empiriolabs/deepseek-v4-pro-el": { - "id": "empiriolabs/deepseek-v4-pro-el", - "name": "DeepSeek-V4-Pro-EL", + "duo-chat-gpt-5-4-mini": { + "id": "duo-chat-gpt-5-4-mini", + "name": "Agentic Chat (GPT-5.4 Mini)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2026-04-24", - "last_updated": "2026-05-02", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 384000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1.67, - "output": 3.33 + "input": 0, + "output": 0 } }, - "empiriolabs/deepseek-v4-flash-el": { - "id": "empiriolabs/deepseek-v4-flash-el", - "name": "DeepSeek-V4-Flash-EL", + "duo-chat-gpt-5-3-codex": { + "id": "duo-chat-gpt-5-3-codex", + "name": "Agentic Chat (GPT-5.3 Codex)", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2026-04-24", - "last_updated": "2026-05-02", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "input": 1000000, - "output": 384000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.28 + "input": 0, + "output": 0 } }, - "elevenlabs/elevenlabs-v2.5-turbo": { - "id": "elevenlabs/elevenlabs-v2.5-turbo", - "name": "ElevenLabs-v2.5-Turbo", - "family": "elevenlabs", + "duo-chat-haiku-4-5": { + "id": "duo-chat-haiku-4-5", + "name": "Agentic Chat (Claude Haiku 4.5)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2024-10-28", - "last_updated": "2024-10-28", + "temperature": true, + "knowledge": "2025-02-28", + "release_date": "2026-01-08", + "last_updated": "2026-01-08", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 0 + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "elevenlabs/elevenlabs-v3": { - "id": "elevenlabs/elevenlabs-v3", - "name": "ElevenLabs-v3", - "family": "elevenlabs", + "duo-chat-gpt-5-2": { + "id": "duo-chat-gpt-5-2", + "name": "Agentic Chat (GPT-5.2)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "knowledge": "2025-08-31", + "release_date": "2026-01-23", + "last_updated": "2026-01-23", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0, "output": 0 } }, - "elevenlabs/elevenlabs-music": { - "id": "elevenlabs/elevenlabs-music", - "name": "ElevenLabs-Music", - "family": "elevenlabs", + "duo-chat-sonnet-4-5": { + "id": "duo-chat-sonnet-4-5", + "name": "Agentic Chat (Claude Sonnet 4.5)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-08-29", - "last_updated": "2025-08-29", + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2026-01-08", + "last_updated": "2026-01-08", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000, - "output": 0 + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "cerebras/gpt-oss-120b-cs": { - "id": "cerebras/gpt-oss-120b-cs", - "name": "GPT-OSS-120B-CS", + "duo-chat-gpt-5-1": { + "id": "duo-chat-gpt-5-1", + "name": "Agentic Chat (GPT-5.1)", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "knowledge": "2024-09-30", + "release_date": "2026-01-22", + "last_updated": "2026-01-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 0 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.35, - "output": 0.75 + "input": 0, + "output": 0 } - }, - "cerebras/llama-3.1-8b-cs": { - "id": "cerebras/llama-3.1-8b-cs", - "name": "Llama-3.1-8B-CS", - "attachment": true, + } + } + }, + "kilo": { + "id": "kilo", + "env": ["KILO_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.kilo.ai/api/gateway", + "name": "Kilo Gateway", + "doc": "https://kilo.ai", + "models": { + "inclusionai/ling-2.6-1t": { + "id": "inclusionai/ling-2.6-1t", + "name": "inclusionAI: Ling-2.6-1T", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-05-13", - "last_updated": "2025-05-13", + "temperature": true, + "release_date": "2026-04-23", + "last_updated": "2026-05-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 0 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.3, + "output": 2.5, + "cache_read": 0.06 } }, - "cerebras/qwen3-32b-cs": { - "id": "cerebras/qwen3-32b-cs", - "name": "qwen3-32b-cs", - "attachment": true, + "inclusionai/ring-2.6-1t": { + "id": "inclusionai/ring-2.6-1t", + "name": "inclusionAI: Ring-2.6-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-05-15", - "last_updated": "2025-05-15", + "temperature": true, + "release_date": "2026-05-08", + "last_updated": "2026-05-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 262144, + "output": 65536 }, - "status": "deprecated" + "cost": { + "input": 0.075, + "output": 0.625, + "cache_read": 0.015 + } }, - "cerebras/qwen3-235b-2507-cs": { - "id": "cerebras/qwen3-235b-2507-cs", - "name": "qwen3-235b-2507-cs", - "attachment": true, - "reasoning": true, + "inclusionai/ling-2.6-flash": { + "id": "inclusionai/ling-2.6-flash", + "name": "inclusionAI: Ling-2.6 Flash", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-08-06", - "last_updated": "2025-08-06", + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-05-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 262144, + "output": 32768 }, - "status": "deprecated" + "cost": { + "input": 0.08, + "output": 0.24, + "cache_read": 0.016 + } }, - "cerebras/llama-3.3-70b-cs": { - "id": "cerebras/llama-3.3-70b-cs", - "name": "llama-3.3-70b-cs", - "attachment": true, + "ibm-granite/granite-4.0-h-micro": { + "id": "ibm-granite/granite-4.0-h-micro", + "name": "IBM: Granite 4.0 Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-05-13", - "last_updated": "2025-05-13", + "temperature": true, + "release_date": "2025-10-20", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 131000, + "output": 32768 }, - "status": "deprecated" + "cost": { + "input": 0.017, + "output": 0.11 + } }, - "stabilityai/stablediffusionxl": { - "id": "stabilityai/stablediffusionxl", - "name": "StableDiffusionXL", - "family": "stable-diffusion", - "attachment": true, + "ibm-granite/granite-4.1-8b": { + "id": "ibm-granite/granite-4.1-8b", + "name": "IBM: Granite 4.1 8B", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2023-07-09", - "last_updated": "2023-07-09", + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 200, - "output": 0 + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.05, + "output": 0.1, + "cache_read": 0.05 } }, - "xai/grok-code-fast-1": { - "id": "xai/grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", - "attachment": true, - "reasoning": true, + "meta-llama/llama-3.1-8b-instruct": { + "id": "meta-llama/llama-3.1-8b-instruct", + "name": "Meta: Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-08-22", - "last_updated": "2025-08-22", + "temperature": true, + "release_date": "2024-07-23", + "last_updated": "2025-12-23", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 16384, + "output": 16384 + }, + "cost": { + "input": 0.02, + "output": 0.05 + } + }, + "meta-llama/llama-3-70b-instruct": { + "id": "meta-llama/llama-3-70b-instruct", + "name": "Meta: Llama 3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 8192, + "output": 8000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.51, + "output": 0.74 } }, - "xai/grok-4-fast-reasoning": { - "id": "xai/grok-4-fast-reasoning", - "name": "Grok-4-Fast-Reasoning", - "family": "grok", - "attachment": true, - "reasoning": true, + "meta-llama/llama-3.1-70b-instruct": { + "id": "meta-llama/llama-3.1-70b-instruct", + "name": "Meta: Llama 3.1 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-09-16", - "last_updated": "2025-09-16", + "temperature": true, + "release_date": "2024-07-16", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 128000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.4, + "output": 0.4 } }, - "xai/grok-4.1-fast-non-reasoning": { - "id": "xai/grok-4.1-fast-non-reasoning", - "name": "Grok-4.1-Fast-Non-Reasoning", - "family": "grok", - "attachment": true, + "meta-llama/llama-3.2-1b-instruct": { + "id": "meta-llama/llama-3.2-1b-instruct", + "name": "Meta: Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "tool_call": false, + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2026-01-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 60000, + "output": 12000 + }, + "cost": { + "input": 0.027, + "output": 0.2 } }, - "xai/grok-4": { - "id": "xai/grok-4", - "name": "Grok-4", - "family": "grok", + "meta-llama/llama-4-maverick": { + "id": "meta-llama/llama-4-maverick", + "name": "Meta: Llama 4 Maverick", + "description": "Open multimodal Llama model for strong reasoning and fast responses", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "temperature": true, + "release_date": "2025-04-05", + "last_updated": "2025-12-24", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.15, + "output": 0.6 } }, - "xai/grok-3-mini": { - "id": "xai/grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", + "meta-llama/llama-3.2-11b-vision-instruct": { + "id": "meta-llama/llama-3.2-11b-vision-instruct", + "name": "Meta: Llama 3.2 11B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 8192 + "output": 16384 }, "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075 + "input": 0.049, + "output": 0.049 } }, - "xai/grok-4.20-multi-agent": { - "id": "xai/grok-4.20-multi-agent", - "name": "Grok-4.20-Multi-Agent", - "attachment": true, + "meta-llama/llama-3.3-70b-instruct": { + "id": "meta-llama/llama-3.3-70b-instruct", + "name": "Meta: Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2026-03-13", - "last_updated": "2026-03-13", + "temperature": true, + "release_date": "2024-08-01", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 0 + "context": 131072, + "output": 16384 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2 + "input": 0.1, + "output": 0.32 } }, - "xai/grok-3": { - "id": "xai/grok-3", - "name": "Grok 3", - "family": "grok", - "attachment": true, + "meta-llama/llama-guard-3-8b": { + "id": "meta-llama/llama-guard-3-8b", + "name": "Llama Guard 3 8B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-11", - "last_updated": "2025-04-11", + "tool_call": false, + "temperature": true, + "release_date": "2024-04-18", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 8192 + "output": 26215 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.02, + "output": 0.06 } }, - "xai/grok-4-fast-non-reasoning": { - "id": "xai/grok-4-fast-non-reasoning", - "name": "Grok-4-Fast-Non-Reasoning", - "family": "grok", + "meta-llama/llama-guard-4-12b": { + "id": "meta-llama/llama-guard-4-12b", + "name": "Meta: Llama Guard 4 12B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-09-16", - "last_updated": "2025-09-16", + "tool_call": false, + "temperature": true, + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 128000 + "context": 163840, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.18, + "output": 0.18 } }, - "xai/grok-4.1-fast-reasoning": { - "id": "xai/grok-4.1-fast-reasoning", - "name": "Grok-4.1-Fast-Reasoning", - "family": "grok", - "attachment": true, - "reasoning": true, + "meta-llama/llama-3-8b-instruct": { + "id": "meta-llama/llama-3-8b-instruct", + "name": "Meta: Llama 3 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "temperature": true, + "release_date": "2024-04-25", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 8192, + "output": 16384 + }, + "cost": { + "input": 0.03, + "output": 0.04 } }, - "runwayml/runway": { - "id": "runwayml/runway", - "name": "Runway", - "family": "runway", + "meta-llama/llama-4-scout": { + "id": "meta-llama/llama-4-scout", + "name": "Meta: Llama 4 Scout", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-10-11", - "last_updated": "2024-10-11", + "temperature": true, + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], - "output": ["video"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256, - "output": 0 + "context": 327680, + "output": 16384 + }, + "cost": { + "input": 0.08, + "output": 0.3 } }, - "runwayml/runway-gen-4-turbo": { - "id": "runwayml/runway-gen-4-turbo", - "name": "Runway-Gen-4-Turbo", - "family": "runway", - "attachment": true, + "meta-llama/llama-3.2-3b-instruct": { + "id": "meta-llama/llama-3.2-3b-instruct", + "name": "Meta: Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-05-09", - "last_updated": "2025-05-09", + "tool_call": false, + "temperature": true, + "release_date": "2024-09-18", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["video"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256, - "output": 0 + "context": 80000, + "output": 16384 + }, + "cost": { + "input": 0.051, + "output": 0.34 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "GPT-5.1-Codex-Max", + "~anthropic/claude-haiku-latest": { + "id": "~anthropic/claude-haiku-latest", + "name": "Anthropic: Claude Haiku Latest", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "openai/sora-2-pro": { - "id": "openai/sora-2-pro", - "name": "Sora-2-Pro", - "family": "sora", + "~anthropic/claude-sonnet-latest": { + "id": "~anthropic/claude-sonnet-latest", + "name": "Anthropic: Claude Sonnet Latest", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], - "output": ["video"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openai/chatgpt-4o-latest": { - "id": "openai/chatgpt-4o-latest", - "name": "ChatGPT-4o-Latest", - "family": "gpt", + "~anthropic/claude-opus-latest": { + "id": "~anthropic/claude-opus-latest", + "name": "Anthropic: Claude Opus Latest", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2024-08-14", - "last_updated": "2024-08-14", + "release_date": "2026-04-16", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 128000 }, - "status": "deprecated", "cost": { - "input": 4.5, - "output": 14 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "openai/gpt-5-chat": { - "id": "openai/gpt-5-chat", - "name": "GPT-5-Chat", - "family": "gpt-codex", - "attachment": true, - "reasoning": false, + "kilo-auto/balanced": { + "id": "kilo-auto/balanced", + "name": "Kilo Auto Balanced", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 0.6, + "output": 3 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT-5.2-Pro", + "kilo-auto/small": { + "id": "kilo-auto/small", + "name": "Kilo Auto Small", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, @@ -61613,1209 +108519,1378 @@ "output": 128000 }, "cost": { - "input": 19, - "output": 150 + "input": 0.05, + "output": 0.4 } }, - "openai/gpt-4o-aug": { - "id": "openai/gpt-4o-aug", - "name": "GPT-4o-Aug", - "family": "gpt", - "attachment": true, - "reasoning": false, + "kilo-auto/free": { + "id": "kilo-auto/free", + "name": "Kilo Auto Free", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2024-11-21", - "last_updated": "2024-11-21", + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 2.2, - "output": 9, - "cache_read": 1.1 + "input": 0, + "output": 0 } }, - "openai/gpt-image-2": { - "id": "openai/gpt-image-2", - "name": "GPT-Image-2", + "kilo-auto/frontier": { + "id": "kilo-auto/frontier", + "name": "Kilo Auto Frontier", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["image", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 5.0505, - "output": 32.3232, - "cache_read": 1.2626 + "input": 5, + "output": 25 } }, - "openai/gpt-4-classic-0314": { - "id": "openai/gpt-4-classic-0314", - "name": "GPT-4-Classic-0314", - "family": "gpt", - "attachment": true, + "moonshotai/kimi-k2": { + "id": "moonshotai/kimi-k2", + "name": "MoonshotAI: Kimi K2 0711", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-08-26", - "last_updated": "2024-08-26", + "temperature": true, + "release_date": "2025-07-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 4096 + "context": 131000, + "output": 26215 }, - "status": "deprecated", "cost": { - "input": 27, - "output": 54 + "input": 0.55, + "output": 2.2 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5-mini", - "family": "gpt-mini", - "attachment": true, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "MoonshotAI: Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-06-25", - "last_updated": "2025-06-25", + "temperature": true, + "release_date": "2025-11-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 65535 }, "cost": { - "input": 0.22, - "output": 1.8, - "cache_read": 0.022 + "input": 0.47, + "output": 2, + "cache_read": 0.2 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT-5-nano", - "family": "gpt-nano", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "MoonshotAI: Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 65535 }, "cost": { - "input": 0.045, - "output": 0.36, - "cache_read": 0.0045 + "input": 0.45, + "output": 2.2 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT-5.3-Codex", + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "MoonshotAI: Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-02-10", - "last_updated": "2026-02-10", + "structured_output": true, + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-05-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 65535 }, "cost": { - "input": 1.6, - "output": 13, - "cache_read": 0.16 + "input": 0.75, + "output": 3.5, + "cache_read": 0.375 } }, - "openai/gpt-4-turbo": { - "id": "openai/gpt-4-turbo", - "name": "GPT-4-Turbo", - "family": "gpt", - "attachment": true, + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "MoonshotAI: Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2023-09-13", - "last_updated": "2023-09-13", + "temperature": true, + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 26215 }, "cost": { - "input": 9, - "output": 27 + "input": 0.4, + "output": 2, + "cache_read": 0.15 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "baidu/ernie-4.5-300b-a47b": { + "id": "baidu/ernie-4.5-300b-a47b", + "name": "Baidu: ERNIE 4.5 300B A47B ", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 123000, + "output": 12000 }, "cost": { - "input": 1.6, - "output": 13, - "cache_read": 0.16 + "input": 0.28, + "output": 1.1 } }, - "openai/o3-pro": { - "id": "openai/o3-pro", - "name": "o3-pro", - "family": "o-pro", + "baidu/ernie-4.5-vl-28b-a3b": { + "id": "baidu/ernie-4.5-vl-28b-a3b", + "name": "Baidu: ERNIE 4.5 VL 28B A3B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-06-10", - "last_updated": "2025-06-10", + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 30000, + "output": 8000 }, "cost": { - "input": 18, - "output": 72 + "input": 0.14, + "output": 0.56 } }, - "openai/o3-mini-high": { - "id": "openai/o3-mini-high", - "name": "o3-mini-high", - "family": "o-mini", + "baidu/ernie-4.5-vl-424b-a47b": { + "id": "baidu/ernie-4.5-vl-424b-a47b", + "name": "Baidu: ERNIE 4.5 VL 424B A47B ", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 123000, + "output": 16000 }, "cost": { - "input": 0.99, - "output": 4 + "input": 0.42, + "output": 1.25 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o-mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "baidu/ernie-4.5-21b-a3b-thinking": { + "id": "baidu/ernie-4.5-21b-a3b-thinking", + "name": "Baidu: ERNIE 4.5 21B A3B Thinking", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 124096, - "output": 4096 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.14, - "output": 0.54, - "cache_read": 0.068 + "input": 0.07, + "output": 0.28 } }, - "openai/o4-mini-deep-research": { - "id": "openai/o4-mini-deep-research", - "name": "o4-mini-deep-research", - "family": "o-mini", - "attachment": true, + "baidu/cobuddy:free": { + "id": "baidu/cobuddy:free", + "name": "Baidu: CoBuddy (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": false, - "release_date": "2025-06-27", - "last_updated": "2025-06-27", + "release_date": "2026-05-06", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 65536 }, "cost": { - "input": 1.8, - "output": 7.2, - "cache_read": 0.45 + "input": 0, + "output": 0 } }, - "openai/gpt-5.4-mini": { - "id": "openai/gpt-5.4-mini", - "name": "GPT-5.4-Mini", - "attachment": true, - "reasoning": true, + "baidu/ernie-4.5-21b-a3b": { + "id": "baidu/ernie-4.5-21b-a3b", + "name": "Baidu: ERNIE 4.5 21B A3B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2026-03-12", - "last_updated": "2026-03-12", + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 120000, + "output": 8000 }, "cost": { - "input": 0.68, - "output": 4, - "cache_read": 0.068 + "input": 0.07, + "output": 0.28 } }, - "openai/dall-e-3": { - "id": "openai/dall-e-3", - "name": "DALL-E-3", - "family": "dall-e", + "baidu/qianfan-ocr-fast": { + "id": "baidu/qianfan-ocr-fast", + "name": "Baidu: Qianfan-OCR-Fast", + "description": "OCR model for extracting structured text from documents and screenshots", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2023-11-06", - "last_updated": "2023-11-06", - "modalities": { - "input": ["text"], - "output": ["image"] - }, - "open_weights": false, - "limit": { - "context": 800, - "output": 0 - } - }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "o4-mini", - "family": "o-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "tool_call": false, + "temperature": true, + "release_date": "2026-04-20", + "last_updated": "2026-05-16", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 65536, + "output": 28672 }, "cost": { - "input": 0.99, - "output": 4, - "cache_read": 0.25 + "input": 0.68, + "output": 2.81 } }, - "openai/gpt-5.4-nano": { - "id": "openai/gpt-5.4-nano", - "name": "GPT-5.4-Nano", + "perceptron/perceptron-mk1": { + "id": "perceptron/perceptron-mk1", + "name": "Perceptron: Perceptron Mk1", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-16", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.18, - "output": 1.1, - "cache_read": 0.018 + "input": 0.15, + "output": 1.5 } }, - "openai/gpt-image-1": { - "id": "openai/gpt-image-1", - "name": "GPT-Image-1", - "family": "gpt", - "attachment": true, + "alfredpros/codellama-7b-instruct-solidity": { + "id": "alfredpros/codellama-7b-instruct-solidity", + "name": "AlfredPros: CodeLLaMa 7B Instruct Solidity", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-03-31", - "last_updated": "2025-03-31", + "tool_call": false, + "temperature": true, + "release_date": "2025-04-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 0 + "context": 4096, + "output": 4096 + }, + "cost": { + "input": 0.8, + "output": 1.2 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2-Codex", + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Google: Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "temperature": true, + "release_date": "2026-05-07", + "last_updated": "2026-05-16", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.6, - "output": 13, - "cache_read": 0.16 + "input": 0.25, + "output": 1.5, + "reasoning": 1.5, + "cache_read": 0.025, + "cache_write": 0.08333 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT-5.1-Codex-Mini", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-11-12", - "last_updated": "2025-11-12", + "google/gemma-3n-e4b-it": { + "id": "google/gemma-3n-e4b-it", + "name": "Google: Gemma 3n 4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 32768, + "output": 6554 }, "cost": { - "input": 0.22, - "output": 1.8, - "cache_read": 0.022 + "input": 0.02, + "output": 0.04 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Google: Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-11-12", - "last_updated": "2025-11-12", + "temperature": true, + "release_date": "2025-03-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 - } - }, - "openai/gpt-image-1-mini": { - "id": "openai/gpt-image-1-mini", - "name": "GPT-Image-1-Mini", - "family": "gpt", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-08-26", - "last_updated": "2025-08-26", - "modalities": { - "input": ["text", "image"], - "output": ["image"] - }, - "open_weights": false, - "limit": { - "context": 0, - "output": 0 + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "openai/o1": { - "id": "openai/o1", - "name": "o1", - "family": "o", + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Google: Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "temperature": false, - "release_date": "2024-12-18", - "last_updated": "2024-12-18", + "temperature": true, + "release_date": "2025-07-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 14, - "output": 54 + "input": 0.3, + "output": 2.5, + "reasoning": 2.5, + "cache_read": 0.03, + "cache_write": 0.083333 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "GPT-5.4-Pro", + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Google: Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-05-19", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["audio", "image", "pdf", "text", "video"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 27, - "output": 160 + "input": 1.5, + "output": 9, + "reasoning": 9, + "cache_read": 0.15, + "cache_write": 0.08333 } }, - "openai/gpt-3.5-turbo": { - "id": "openai/gpt-3.5-turbo", - "name": "GPT-3.5-Turbo", - "family": "gpt", + "google/gemini-2.0-flash-lite-001": { + "id": "google/gemini-2.0-flash-lite-001", + "name": "Google: Gemini 2.0 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2023-09-13", - "last_updated": "2023-09-13", + "temperature": true, + "release_date": "2024-12-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 2048 + "context": 1048576, + "output": 8192 }, "cost": { - "input": 0.45, - "output": 1.4 + "input": 0.075, + "output": 0.3 } }, - "openai/o3-deep-research": { - "id": "openai/o3-deep-research", - "name": "o3-deep-research", - "family": "o", + "google/gemini-2.5-flash-lite-preview-09-2025": { + "id": "google/gemini-2.5-flash-lite-preview-09-2025", + "name": "Google: Gemini 2.5 Flash Lite Preview 09-2025", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-06-27", - "last_updated": "2025-06-27", + "temperature": true, + "release_date": "2025-09-25", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 9, - "output": 36, - "cache_read": 2.2 + "input": 0.1, + "output": 0.4, + "reasoning": 0.4, + "cache_read": 0.01, + "cache_write": 0.083333 } }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "o3-mini", - "family": "o-mini", + "google/gemini-2.0-flash-001": { + "id": "google/gemini-2.0-flash-001", + "name": "Google: Gemini 2.0 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "temperature": true, + "release_date": "2024-12-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 8192 }, "cost": { - "input": 0.99, - "output": 4 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025, + "cache_write": 0.083333 } }, - "openai/o1-pro": { - "id": "openai/o1-pro", - "name": "o1-pro", - "family": "o-pro", + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Google: Gemma 4 31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-03-19", - "last_updated": "2025-03-19", + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 140, - "output": 540 + "input": 0.14, + "output": 0.4 } }, - "openai/gpt-4o-search": { - "id": "openai/gpt-4o-search", - "name": "GPT-4o-Search", - "family": "gpt", + "google/lyria-3-clip-preview": { + "id": "google/lyria-3-clip-preview", + "name": "Google: Lyria 3 Clip Preview", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-03-11", - "last_updated": "2025-03-11", + "tool_call": false, + "temperature": true, + "release_date": "2026-03-30", + "last_updated": "2026-04-11", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["image", "text"], + "output": ["audio", "text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 2.2, - "output": 9 + "input": 0, + "output": 0 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5-Codex", - "family": "gpt-codex", + "google/gemini-3.1-pro-preview-customtools": { + "id": "google/gemini-3.1-pro-preview-customtools", + "name": "Google: Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 9 + "input": 2, + "output": 12, + "reasoning": 12 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", + "google/gemini-3-pro-image-preview": { + "id": "google/gemini-3-pro-image-preview", + "name": "Google: Nano Banana Pro (Gemini 3 Pro Image Preview)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": false, + "temperature": true, + "release_date": "2025-11-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["image"] + "input": ["image", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 65536, + "output": 32768 }, "cost": { - "input": 2.2, - "output": 14, - "cache_read": 0.22 + "input": 2, + "output": 12, + "reasoning": 12 } }, - "openai/gpt-5.3-codex-spark": { - "id": "openai/gpt-5.3-codex-spark", - "name": "GPT-5.3-Codex-Spark", + "google/gemini-2.5-flash-image": { + "id": "google/gemini-2.5-flash-image", + "name": "Google: Nano Banana (Gemini 2.5 Flash Image)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-03-04", - "last_updated": "2026-03-04", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-10-08", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["image", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 2.5 } }, - "openai/gpt-3.5-turbo-raw": { - "id": "openai/gpt-3.5-turbo-raw", - "name": "GPT-3.5-Turbo-Raw", - "family": "gpt", + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Google: Gemini 2.5 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "temperature": false, - "release_date": "2023-09-27", - "last_updated": "2023-09-27", + "temperature": true, + "release_date": "2025-06-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4524, - "output": 2048 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 0.45, - "output": 1.4 + "input": 0.1, + "output": 0.4, + "reasoning": 0.4, + "cache_read": 0.01, + "cache_write": 0.083333 } }, - "openai/gpt-4.1-nano": { - "id": "openai/gpt-4.1-nano", - "name": "GPT-4.1-nano", - "family": "gpt-nano", + "google/gemini-3.1-flash-image-preview": { + "id": "google/gemini-3.1-flash-image-preview", + "name": "Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": false, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["image", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 65536, + "output": 65536 }, "cost": { - "input": 0.09, - "output": 0.36, - "cache_read": 0.022 + "input": 0.5, + "output": 3 } }, - "openai/o3": { - "id": "openai/o3", - "name": "o3", - "family": "o", + "google/gemini-2.5-pro-preview-05-06": { + "id": "google/gemini-2.5-pro-preview-05-06", + "name": "Google: Gemini 2.5 Pro Preview 05-06", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "release_date": "2025-05-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 1.8, - "output": 7.2, - "cache_read": 0.45 + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "GPT-5-Pro", - "family": "gpt-pro", + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Google: Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "release_date": "2026-02-19", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 14, - "output": 110 + "input": 2, + "output": 12, + "reasoning": 12 } }, - "openai/sora-2": { - "id": "openai/sora-2", - "name": "Sora-2", - "family": "sora", + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Google: Gemma 4 26B A4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "release_date": "2026-04-03", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], - "output": ["video"] + "input": ["image", "text", "video"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.12, + "output": 0.4 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "GPT-4o", - "family": "gpt", + "google/gemini-2.5-pro-preview": { + "id": "google/gemini-2.5-pro-preview", + "name": "Google: Gemini 2.5 Pro Preview 06-05", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "temperature": true, + "release_date": "2025-06-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", - "family": "gpt", + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Google: Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "temperature": true, + "release_date": "2025-12-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 0.5, + "output": 3, + "reasoning": 3, + "cache_read": 0.05, + "cache_write": 0.083333 } }, - "openai/gpt-5.2-instant": { - "id": "openai/gpt-5.2-instant", - "name": "GPT-5.2-Instant", + "google/gemma-3-12b-it": { + "id": "google/gemma-3-12b-it", + "name": "Google: Gemma 3 12B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "tool_call": false, + "temperature": true, + "release_date": "2025-03-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1.6, - "output": 13, - "cache_read": 0.16 + "input": 0.04, + "output": 0.13, + "cache_read": 0.015 } }, - "openai/gpt-4o-mini-search": { - "id": "openai/gpt-4o-mini-search", - "name": "GPT-4o-mini-Search", - "family": "gpt-mini", + "google/gemma-3-4b-it": { + "id": "google/gemma-3-4b-it", + "name": "Google: Gemma 3 4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-03-11", - "last_updated": "2025-03-11", + "tool_call": false, + "temperature": true, + "release_date": "2025-03-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 19200 }, "cost": { - "input": 0.14, - "output": 0.54 + "input": 0.04, + "output": 0.08 } }, - "openai/gpt-image-1.5": { - "id": "openai/gpt-image-1.5", - "name": "gpt-image-1.5", + "google/gemma-3-27b-it": { + "id": "google/gemma-3-27b-it", + "name": "Google: Gemma 3 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "tool_call": true, + "temperature": true, + "release_date": "2025-03-12", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["image", "text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 0 + "output": 65536 + }, + "cost": { + "input": 0.03, + "output": 0.11, + "cache_read": 0.02 } }, - "openai/gpt-3.5-turbo-instruct": { - "id": "openai/gpt-3.5-turbo-instruct", - "name": "GPT-3.5-Turbo-Instruct", - "family": "gpt", + "google/lyria-3-pro-preview": { + "id": "google/lyria-3-pro-preview", + "name": "Google: Lyria 3 Pro Preview", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2023-09-20", - "last_updated": "2023-09-20", + "tool_call": false, + "temperature": true, + "release_date": "2026-03-30", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["image", "text"], + "output": ["audio", "text"] }, "open_weights": false, "limit": { - "context": 3500, - "output": 1024 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.4, - "output": 1.8 + "input": 0, + "output": 0 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "google/gemma-2-27b-it": { + "id": "google/gemma-2-27b-it", + "name": "Google: Gemma 2 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "temperature": true, + "release_date": "2024-06-24", + "last_updated": "2024-06-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 8192, + "output": 2048 }, "cost": { - "input": 1.8, - "output": 7.2, - "cache_read": 0.45 + "input": 0.65, + "output": 0.65 } }, - "openai/gpt-5.1-instant": { - "id": "openai/gpt-5.1-instant", - "name": "GPT-5.1-Instant", - "family": "gpt", + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Google: Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-11-12", - "last_updated": "2025-11-12", + "temperature": true, + "release_date": "2026-03-03", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["audio", "image", "pdf", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 0.25, + "output": 1.5, + "reasoning": 1.5 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT-4.1-mini", - "family": "gpt-mini", - "attachment": true, + "liquid/lfm-2-24b-a2b": { + "id": "liquid/lfm-2-24b-a2b", + "name": "LiquidAI: LFM2-24B-A2B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "tool_call": false, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, + "context": 32768, "output": 32768 }, "cost": { - "input": 0.36, - "output": 1.4, - "cache_read": 0.09 + "input": 0.03, + "output": 0.12 } }, - "openai/gpt-4-classic": { - "id": "openai/gpt-4-classic", - "name": "GPT-4-Classic", - "family": "gpt", + "x-ai/grok-4.20": { + "id": "x-ai/grok-4.20", + "name": "xAI: Grok 4.20", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "release_date": "2024-03-25", - "last_updated": "2024-03-25", + "temperature": true, + "release_date": "2026-03-31", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 4096 + "context": 2000000, + "output": 2000000 }, - "status": "deprecated", "cost": { - "input": 27, - "output": 54 + "input": 2, + "output": 6, + "cache_read": 0.2 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1-Codex", - "family": "gpt-codex", + "x-ai/grok-4.3": { + "id": "x-ai/grok-4.3", + "name": "xAI: Grok 4.3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-11-12", - "last_updated": "2025-11-12", + "temperature": true, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 4096 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "openai/gpt-5.3-instant": { - "id": "openai/gpt-5.3-instant", - "name": "GPT-5.3-Instant", + "x-ai/grok-4.20-multi-agent": { + "id": "x-ai/grok-4.20-multi-agent", + "name": "xAI: Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": false, + "temperature": true, + "release_date": "2026-03-31", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 1.6, - "output": 13, - "cache_read": 0.16 + "input": 2, + "output": 6, + "cache_read": 0.2 } }, - "google/veo-3-fast": { - "id": "google/veo-3-fast", - "name": "Veo-3-Fast", - "family": "veo", + "x-ai/grok-build-0.1": { + "id": "x-ai/grok-build-0.1", + "name": "xAI: Grok Build 0.1", + "description": "Grok coding model for agentic engineering, edits, and codebase workflows", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-10-13", - "last_updated": "2025-10-13", + "temperature": true, + "release_date": "2026-05-20", + "last_updated": "2026-05-27", "modalities": { - "input": ["text"], - "output": ["video"] + "input": ["image", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 1, + "output": 2, + "cache_read": 0.2 } }, - "google/veo-3.1-fast": { - "id": "google/veo-3.1-fast", - "name": "Veo-3.1-Fast", - "family": "veo", + "~google/gemini-pro-latest": { + "id": "~google/gemini-pro-latest", + "name": "Google: Gemini Pro Latest", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], - "output": ["video"] + "input": ["text", "image", "audio", "video", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2, + "cache_write": 0.375 } }, - "google/gemini-3.1-pro": { - "id": "google/gemini-3.1-pro", - "name": "Gemini-3.1-Pro", + "~google/gemini-flash-latest": { + "id": "~google/gemini-flash-latest", + "name": "Google: Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, @@ -62824,1000 +109899,1216 @@ "output": 65536 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.08333333333333334 } }, - "google/imagen-3-fast": { - "id": "google/imagen-3-fast", - "name": "Imagen-3-Fast", - "family": "imagen", - "attachment": true, + "microsoft/phi-4-mini-instruct": { + "id": "microsoft/phi-4-mini-instruct", + "name": "Microsoft: Phi 4 Mini Instruct", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2024-10-17", - "last_updated": "2024-10-17", + "tool_call": false, + "structured_output": true, + "temperature": true, + "release_date": "2025-10-17", + "last_updated": "2026-05-07", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 128000, + "output": 128000 + }, + "cost": { + "input": 0.08, + "output": 0.35, + "cache_read": 0.08 } }, - "google/gemini-2.0-flash": { - "id": "google/gemini-2.0-flash", - "name": "Gemini-2.0-Flash", - "family": "gemini-flash", - "attachment": true, + "microsoft/phi-4": { + "id": "microsoft/phi-4", + "name": "Microsoft: Phi 4", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, + "tool_call": false, + "temperature": true, "release_date": "2024-12-11", "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 990000, - "output": 8192 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.42 + "input": 0.06, + "output": 0.14 } }, - "google/gemini-deep-research": { - "id": "google/gemini-deep-research", - "name": "gemini-deep-research", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "microsoft/wizardlm-2-8x22b": { + "id": "microsoft/wizardlm-2-8x22b", + "name": "WizardLM-2 8x22B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-04-24", + "last_updated": "2024-04-24", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 0 + "context": 65535, + "output": 8000 }, - "status": "deprecated", "cost": { - "input": 1.6, - "output": 9.6 + "input": 0.62, + "output": 0.62 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini-2.5-Pro", - "family": "gemini-pro", - "attachment": true, + "poolside/laguna-xs.2:free": { + "id": "poolside/laguna-xs.2:free", + "name": "Poolside: Laguna XS.2 (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "laguna", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2025-02-05", - "last_updated": "2025-02-05", + "structured_output": false, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1065535, - "output": 65535 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.87, - "output": 7, - "cache_read": 0.087 + "input": 0, + "output": 0 } }, - "google/imagen-3": { - "id": "google/imagen-3", - "name": "Imagen-3", - "family": "imagen", - "attachment": true, - "reasoning": false, + "poolside/laguna-m.1:free": { + "id": "poolside/laguna-m.1:free", + "name": "Poolside: Laguna M.1 (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "laguna", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "release_date": "2024-10-15", - "last_updated": "2024-10-15", + "structured_output": false, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, "output": 0 } }, - "google/nano-banana": { - "id": "google/nano-banana", - "name": "Nano-Banana", - "family": "nano-banana", - "attachment": true, + "writer/palmyra-x5": { + "id": "writer/palmyra-x5", + "name": "Writer: Palmyra X5", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "tool_call": false, + "temperature": true, + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 0 + "context": 1040000, + "output": 8192 }, "cost": { - "input": 0.21, - "output": 1.8, - "cache_read": 0.021 + "input": 0.6, + "output": 6 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini-2.5-Flash", - "family": "gemini-flash", - "attachment": true, + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "Z.ai: GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-04-26", - "last_updated": "2025-04-26", + "temperature": true, + "release_date": "2025-12-22", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1065535, + "context": 202752, "output": 65535 }, "cost": { - "input": 0.21, - "output": 1.8, - "cache_read": 0.021 + "input": 0.38, + "output": 1.98, + "cache_read": 0.2 } }, - "google/gemini-3.1-flash-lite": { - "id": "google/gemini-3.1-flash-lite", - "name": "Gemini-3.1-Flash-Lite", + "z-ai/glm-4.5v": { + "id": "z-ai/glm-4.5v", + "name": "Z.ai: GLM 4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2026-02-18", - "last_updated": "2026-02-18", + "temperature": true, + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 1.5 + "input": 0.6, + "output": 1.8, + "cache_read": 0.11 } }, - "google/gemini-3-flash": { - "id": "google/gemini-3-flash", - "name": "Gemini-3-Flash", - "attachment": true, + "z-ai/glm-4.5": { + "id": "z-ai/glm-4.5", + "name": "Z.ai: GLM 4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-07", - "last_updated": "2025-10-07", + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.4, - "output": 2.4, - "cache_read": 0.04 + "input": 0.6, + "output": 2.2, + "cache_read": 0.175 } }, - "google/veo-3.1": { - "id": "google/veo-3.1", - "name": "Veo-3.1", - "family": "veo", - "attachment": true, - "reasoning": false, + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "Z.ai: GLM 5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { "input": ["text"], - "output": ["video"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 202752, + "output": 131072 + }, + "cost": { + "input": 1.26, + "output": 3.96 } }, - "google/lyria": { - "id": "google/lyria", - "name": "Lyria", - "family": "lyria", - "attachment": true, - "reasoning": false, + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "Z.ai: GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-06-04", - "last_updated": "2025-06-04", + "temperature": true, + "release_date": "2025-09-30", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], - "output": ["audio"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 204800, + "output": 204800 + }, + "cost": { + "input": 0.39, + "output": 1.9, + "cache_read": 0.175 } }, - "google/imagen-4-ultra": { - "id": "google/imagen-4-ultra", - "name": "Imagen-4-Ultra", - "family": "imagen", - "attachment": true, + "z-ai/glm-4-32b": { + "id": "z-ai/glm-4-32b", + "name": "Z.ai: GLM 4 32B ", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-05-24", - "last_updated": "2025-05-24", + "temperature": true, + "release_date": "2025-07-25", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 128000, + "output": 32768 + }, + "cost": { + "input": 0.1, + "output": 0.1 } }, - "google/nano-banana-pro": { - "id": "google/nano-banana-pro", - "name": "Nano-Banana-Pro", - "family": "nano-banana", + "z-ai/glm-4.6v": { + "id": "z-ai/glm-4.6v", + "name": "Z.ai: GLM 4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "temperature": true, + "release_date": "2025-09-30", + "last_updated": "2026-01-10", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["image", "text", "video"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 65536, - "output": 0 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2 + "input": 0.3, + "output": 0.9 } }, - "google/gemini-3-pro": { - "id": "google/gemini-3-pro", - "name": "Gemini-3-Pro", - "family": "gemini-pro", + "z-ai/glm-5v-turbo": { + "id": "z-ai/glm-5v-turbo", + "name": "Z.ai: GLM 5V Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-10-22", - "last_updated": "2025-10-22", + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 202752, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 1.6, - "output": 9.6, - "cache_read": 0.16 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "google/imagen-4-fast": { - "id": "google/imagen-4-fast", - "name": "Imagen-4-Fast", - "family": "imagen", - "attachment": true, - "reasoning": false, + "z-ai/glm-4.5-air": { + "id": "z-ai/glm-4.5-air", + "name": "Z.ai: GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-06-25", - "last_updated": "2025-06-25", + "temperature": true, + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 131072, + "output": 98304 + }, + "cost": { + "input": 0.13, + "output": 0.85, + "cache_read": 0.025 } }, - "google/veo-3": { - "id": "google/veo-3", - "name": "Veo-3", - "family": "veo", - "attachment": true, - "reasoning": false, + "z-ai/glm-4.7-flash": { + "id": "z-ai/glm-4.7-flash", + "name": "Z.ai: GLM 4.7 Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-05-21", - "last_updated": "2025-05-21", + "temperature": true, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], - "output": ["video"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 202752, + "output": 40551 + }, + "cost": { + "input": 0.06, + "output": 0.4, + "cache_read": 0.01 } }, - "google/gemini-2.5-flash-lite": { - "id": "google/gemini-2.5-flash-lite", - "name": "Gemini-2.5-Flash-Lite", - "family": "gemini-flash-lite", - "attachment": true, + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "Z.ai: GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-06-19", - "last_updated": "2025-06-19", + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1024000, - "output": 64000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.72, + "output": 2.3 } }, - "google/imagen-4": { - "id": "google/imagen-4", - "name": "Imagen-4", - "family": "imagen", - "attachment": true, - "reasoning": false, + "z-ai/glm-5-turbo": { + "id": "z-ai/glm-5-turbo", + "name": "Z.ai: GLM 5 Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "temperature": true, + "release_date": "2026-03-15", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 480, - "output": 0 + "context": 202752, + "output": 131072 + }, + "cost": { + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "google/gemma-4-31b": { - "id": "google/gemma-4-31b", - "name": "Gemma-4-31B", + "openai/gpt-4o-mini-2024-07-18": { + "id": "openai/gpt-4o-mini-2024-07-18", + "name": "OpenAI: GPT-4o-mini (2024-07-18)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "temperature": true, + "release_date": "2024-07-18", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.6 } }, - "google/gemini-2.0-flash-lite": { - "id": "google/gemini-2.0-flash-lite", - "name": "Gemini-2.0-Flash-Lite", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": false, + "openai/gpt-oss-safeguard-20b": { + "id": "openai/gpt-oss-safeguard-20b", + "name": "OpenAI: gpt-oss-safeguard-20b", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2025-02-05", - "last_updated": "2025-02-05", + "temperature": true, + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 990000, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.052, - "output": 0.21 + "input": 0.075, + "output": 0.3, + "cache_read": 0.037 } }, - "google/veo-2": { - "id": "google/veo-2", - "name": "Veo-2", - "family": "veo", - "attachment": true, + "openai/gpt-3.5-turbo-instruct": { + "id": "openai/gpt-3.5-turbo-instruct", + "name": "OpenAI: GPT-3.5 Turbo Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2024-12-02", - "last_updated": "2024-12-02", + "tool_call": false, + "temperature": true, + "release_date": "2023-03-01", + "last_updated": "2023-09-21", "modalities": { "input": ["text"], - "output": ["video"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 4095, + "output": 4096 + }, + "cost": { + "input": 1.5, + "output": 2 } }, - "lumalabs/ray2": { - "id": "lumalabs/ray2", - "name": "Ray2", - "family": "ray", + "openai/gpt-5.2-chat": { + "id": "openai/gpt-5.2-chat", + "name": "OpenAI: GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, "reasoning": false, "tool_call": true, "temperature": false, - "release_date": "2025-02-20", - "last_updated": "2025-02-20", + "release_date": "2025-12-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["video"] + "input": ["image", "pdf", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 5000, - "output": 0 + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Claude-Opus-4.1", - "family": "claude-opus", + "openai/o3": { + "id": "openai/o3", + "name": "OpenAI: o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-04-16", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 196608, - "output": 32000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 13, - "output": 64, - "cache_read": 1.3, - "cache_write": 16 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "anthropic/claude-sonnet-3.5": { - "id": "anthropic/claude-sonnet-3.5", - "name": "Claude-Sonnet-3.5", - "family": "claude-sonnet", + "openai/o4-mini-high": { + "id": "openai/o4-mini-high", + "name": "OpenAI: o4 Mini High", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "release_date": "2024-06-05", - "last_updated": "2024-06-05", + "release_date": "2025-04-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 189096, - "output": 8192 + "context": 200000, + "output": 100000 }, - "status": "deprecated", "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 1.1, + "output": 4.4 } }, - "anthropic/claude-haiku-3": { - "id": "anthropic/claude-haiku-3", - "name": "Claude-Haiku-3", - "family": "claude-haiku", - "attachment": true, + "openai/gpt-audio": { + "id": "openai/gpt-audio", + "name": "OpenAI: GPT Audio", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": false, - "release_date": "2024-03-09", - "last_updated": "2024-03-09", + "tool_call": false, + "temperature": true, + "release_date": "2026-01-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["audio", "text"], + "output": ["audio", "text"] }, "open_weights": false, "limit": { - "context": 189096, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.21, - "output": 1.1, - "cache_read": 0.021, - "cache_write": 0.26 + "input": 2.5, + "output": 10 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Claude-Opus-4.6", + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "OpenAI: GPT-5.2 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2026-02-04", - "last_updated": "2026-02-04", + "release_date": "2025-12-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 983040, + "context": 400000, "output": 128000 }, "cost": { - "input": 4.3, - "output": 21, - "cache_read": 0.43, - "cache_write": 5.3 + "input": 21, + "output": 168 } }, - "anthropic/claude-opus-4.7": { - "id": "anthropic/claude-opus-4.7", - "name": "Claude-Opus-4.7", - "attachment": true, - "reasoning": true, - "tool_call": true, + "openai/gpt-4o-mini-search-preview": { + "id": "openai/gpt-4o-mini-search-preview", + "name": "OpenAI: GPT-4o-mini Search Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": false, - "release_date": "2026-04-15", - "last_updated": "2026-04-15", + "release_date": "2025-01", + "last_updated": "2025-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 4.3, - "output": 21, - "cache_read": 0.43, - "cache_write": 5.4 + "input": 0.15, + "output": 0.6 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude-Sonnet-4", - "family": "claude-sonnet", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "OpenAI: GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-05-21", - "last_updated": "2025-05-21", + "release_date": "2025-08-07", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 983040, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "anthropic/claude-sonnet-4.5": { - "id": "anthropic/claude-sonnet-4.5", - "name": "Claude-Sonnet-4.5", - "family": "claude-sonnet", + "openai/gpt-5-chat": { + "id": "openai/gpt-5-chat", + "name": "OpenAI: GPT-5 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": false, - "release_date": "2025-09-26", - "last_updated": "2025-09-26", + "release_date": "2025-08-07", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 983040, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "anthropic/claude-opus-4.5": { - "id": "anthropic/claude-opus-4.5", - "name": "Claude-Opus-4.5", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "OpenAI: GPT-3.5 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-21", - "last_updated": "2025-11-21", + "temperature": true, + "release_date": "2023-03-01", + "last_updated": "2023-11-06", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 196608, - "output": 64000 + "context": 16385, + "output": 4096 }, "cost": { - "input": 4.3, - "output": 21, - "cache_read": 0.43, - "cache_write": 5.3 + "input": 0.5, + "output": 1.5 } }, - "anthropic/claude-sonnet-3.7": { - "id": "anthropic/claude-sonnet-3.7", - "name": "Claude-Sonnet-3.7", - "family": "claude-sonnet", + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "OpenAI: GPT-5 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "release_date": "2025-10-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 196608, + "context": 400000, "output": 128000 }, "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 15, + "output": 120 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude-Opus-4", - "family": "claude-opus", + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "OpenAI: GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-05-21", - "last_updated": "2025-05-21", + "temperature": true, + "release_date": "2024-05-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 192512, - "output": 28672 + "context": 128000, + "output": 16384 }, "cost": { - "input": 13, - "output": 64, - "cache_read": 1.3, - "cache_write": 16 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "anthropic/claude-haiku-3.5": { - "id": "anthropic/claude-haiku-3.5", - "name": "Claude-Haiku-3.5", - "family": "claude-haiku", - "attachment": true, + "openai/gpt-4": { + "id": "openai/gpt-4", + "name": "OpenAI: GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "temperature": true, + "release_date": "2023-03-14", + "last_updated": "2024-04-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 189096, - "output": 8192 + "context": 8191, + "output": 4096 }, "cost": { - "input": 0.68, - "output": 3.4, - "cache_read": 0.068, - "cache_write": 0.85 + "input": 30, + "output": 60 } }, - "anthropic/claude-haiku-4.5": { - "id": "anthropic/claude-haiku-4.5", - "name": "Claude-Haiku-4.5", - "family": "claude-haiku", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "OpenAI: o4 Mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2025-04-16", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 192000, - "output": 64000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.85, - "output": 4.3, - "cache_read": 0.085, - "cache_write": 1.1 + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "anthropic/claude-sonnet-3.5-june": { - "id": "anthropic/claude-sonnet-3.5-june", - "name": "Claude-Sonnet-3.5-June", - "family": "claude-sonnet", - "attachment": true, + "openai/gpt-3.5-turbo-16k": { + "id": "openai/gpt-3.5-turbo-16k", + "name": "OpenAI: GPT-3.5 Turbo 16k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-11-18", - "last_updated": "2024-11-18", + "temperature": true, + "release_date": "2023-08-28", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 189096, - "output": 8192 + "context": 16385, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 3, + "output": 4 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Claude-Sonnet-4.6", + "openai/o3-pro": { + "id": "openai/o3-pro", + "name": "OpenAI: o3 Pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2025-04-16", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 983040, - "output": 128000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 2.6, - "output": 13, - "cache_read": 0.26, - "cache_write": 3.2 + "input": 20, + "output": 80 } }, - "ideogramai/ideogram": { - "id": "ideogramai/ideogram", - "name": "Ideogram", - "family": "ideogram", + "openai/gpt-5.1-chat": { + "id": "openai/gpt-5.1-chat", + "name": "OpenAI: GPT-5.1 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, "reasoning": false, "tool_call": true, "temperature": false, - "release_date": "2024-04-03", - "last_updated": "2024-04-03", + "release_date": "2025-11-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["image", "pdf", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 150, - "output": 0 + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "ideogramai/ideogram-v2": { - "id": "ideogramai/ideogram-v2", - "name": "Ideogram-v2", - "family": "ideogram", + "openai/gpt-4o-2024-05-13": { + "id": "openai/gpt-4o-2024-05-13", + "name": "OpenAI: GPT-4o (2024-05-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-08-21", - "last_updated": "2024-08-21", + "temperature": true, + "release_date": "2024-05-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["image", "pdf", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 150, - "output": 0 + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 5, + "output": 15 } }, - "ideogramai/ideogram-v2a-turbo": { - "id": "ideogramai/ideogram-v2a-turbo", - "name": "Ideogram-v2a-Turbo", - "family": "ideogram", - "attachment": true, + "openai/gpt-4-0314": { + "id": "openai/gpt-4-0314", + "name": "OpenAI: GPT-4 (older v0314)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "temperature": true, + "release_date": "2023-05-28", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 150, - "output": 0 + "context": 8191, + "output": 4096 + }, + "cost": { + "input": 30, + "output": 60 } }, - "ideogramai/ideogram-v2a": { - "id": "ideogramai/ideogram-v2a", - "name": "Ideogram-v2a", - "family": "ideogram", + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "OpenAI: GPT-5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": false, - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "release_date": "2026-03-17", + "last_updated": "2026-04-11", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["image", "pdf", "text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 150, - "output": 0 + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "trytako/tako": { - "id": "trytako/tako", - "name": "Tako", - "family": "tako", + "openai/gpt-5.3-chat": { + "id": "openai/gpt-5.3-chat", + "name": "OpenAI: GPT-5.3 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "release_date": "2026-03-04", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2048, - "output": 0 + "context": 128000, + "output": 16384 + }, + "cost": { + "input": 1.75, + "output": 14 } }, - "poetools/claude-code": { - "id": "poetools/claude-code", - "name": "claude-code", - "attachment": true, - "reasoning": true, + "openai/gpt-3.5-turbo-0613": { + "id": "openai/gpt-3.5-turbo-0613", + "name": "OpenAI: GPT-3.5 Turbo (older v0613)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "release_date": "2025-11-27", - "last_updated": "2025-11-27", + "temperature": true, + "release_date": "2023-06-13", + "last_updated": "2023-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 4095, + "output": 4096 + }, + "cost": { + "input": 1, + "output": 2 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "openai/gpt-5-image-mini": { + "id": "openai/gpt-5-image-mini", + "name": "OpenAI: GPT-5 Image Mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "temperature": true, + "release_date": "2025-10-16", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["image", "pdf", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { @@ -63825,26 +111116,29 @@ "output": 128000 }, "cost": { - "input": 4.5455, - "output": 27.2727, - "cache_read": 0.4545 + "input": 2.5, + "output": 2 } }, - "openai/gpt-5.5-pro": { - "id": "openai/gpt-5.5-pro", - "name": "GPT-5.5-Pro", - "family": "gpt-pro", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "OpenAI: GPT-5.1-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { @@ -63852,160 +111146,165 @@ "output": 128000 }, "cost": { - "input": 27.2727, - "output": 163.6364 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } - } - } - }, - "helicone": { - "id": "helicone", - "env": ["HELICONE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://ai-gateway.helicone.ai/v1", - "name": "Helicone", - "doc": "https://helicone.ai/models", - "models": { - "mistral-nemo": { - "id": "mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + }, + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "OpenAI: GPT-5.1-Codex-Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16400 + "context": 400000, + "output": 128000 }, "cost": { - "input": 20, - "output": 40 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "xAI Grok 4.1 Fast Reasoning", - "family": "grok", - "attachment": false, - "reasoning": true, + "openai/gpt-4o-2024-08-06": { + "id": "openai/gpt-4o-2024-08-06", + "name": "OpenAI: GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-17", - "last_updated": "2025-11-17", + "release_date": "2024-08-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.19999999999999998, - "output": 0.5, - "cache_read": 0.049999999999999996 - } - }, - "gemma2-9b-it": { - "id": "gemma2-9b-it", - "name": "Google Gemma 2", - "family": "gemma", - "attachment": false, + "input": 2.5, + "output": 10, + "cache_read": 1.25 + } + }, + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "OpenAI: o3 Mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-25", - "last_updated": "2024-06-25", + "tool_call": true, + "temperature": false, + "release_date": "2024-12-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.01, - "output": 0.03 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Meta Llama 3.3 70B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "OpenAI: GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "temperature": false, + "release_date": "2025-12-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16400 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.13, - "output": 0.39 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "llama-4-scout": { - "id": "llama-4-scout", - "name": "Meta Llama 4 Scout 17B 16E", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "OpenAI: GPT-5.3-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-02-25", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.08, - "output": 0.3 + "input": 1.75, + "output": 14 } }, - "chatgpt-4o-latest": { - "id": "chatgpt-4o-latest", - "name": "OpenAI ChatGPT-4o", - "family": "gpt", + "openai/gpt-audio-mini": { + "id": "openai/gpt-audio-mini", + "name": "OpenAI: GPT Audio Mini", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-14", - "last_updated": "2024-08-14", + "release_date": "2026-01-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["audio", "text"], + "output": ["audio", "text"] }, "open_weights": false, "limit": { @@ -64013,281 +111312,309 @@ "output": 16384 }, "cost": { - "input": 5, - "output": 20, - "cache_read": 2.5 + "input": 0.6, + "output": 2.4 } }, - "claude-3.5-sonnet-v2": { - "id": "claude-3.5-sonnet-v2", - "name": "Anthropic: Claude 3.5 Sonnet v2", - "family": "claude-sonnet", - "attachment": false, - "reasoning": false, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "OpenAI: GPT-5.1-Codex-Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "temperature": false, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 400000, + "output": 100000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.30000000000000004, - "cache_write": 3.75 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "hermes-2-pro-llama-3-8b": { - "id": "hermes-2-pro-llama-3-8b", - "name": "Hermes 2 Pro Llama 3 8B", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/o4-mini-deep-research": { + "id": "openai/o4-mini-deep-research", + "name": "OpenAI: o4 Mini Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-05", - "release_date": "2024-05-27", - "last_updated": "2024-05-27", + "release_date": "2024-06-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.14, - "output": 0.14 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "claude-3.7-sonnet": { - "id": "claude-3.7-sonnet", - "name": "Anthropic: Claude 3.7 Sonnet", - "family": "claude-sonnet", - "attachment": false, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "OpenAI: GPT-4.1 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-02", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "release_date": "2025-04-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.30000000000000004, - "cache_write": 3.75 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "llama-prompt-guard-2-22m": { - "id": "llama-prompt-guard-2-22m", - "name": "Meta Llama Prompt Guard 2 22M", - "family": "llama", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "OpenAI: gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 512, - "output": 2 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.01, - "output": 0.01 + "input": 0.039, + "output": 0.19 } }, - "o1-mini": { - "id": "o1-mini", - "name": "OpenAI: o1-mini", - "family": "o-mini", - "attachment": false, + "openai/gpt-4o-2024-11-20": { + "id": "openai/gpt-4o-2024-11-20", + "name": "OpenAI: GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "tool_call": true, + "temperature": true, + "release_date": "2024-11-20", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 65536 + "output": 16384 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "gpt-4.1-mini-2025-04-14": { - "id": "gpt-4.1-mini-2025-04-14", - "name": "OpenAI GPT-4.1 Mini", - "family": "gpt-mini", - "attachment": false, + "openai/o1": { + "id": "openai/o1", + "name": "OpenAI: o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "temperature": false, + "release_date": "2024-12-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.39999999999999997, - "output": 1.5999999999999999, - "cache_read": 0.09999999999999999 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "deepseek-r1-distill-llama-70b": { - "id": "deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", - "attachment": false, + "openai/o1-pro": { + "id": "openai/o1-pro", + "name": "OpenAI: o1-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": false, + "release_date": "2025-03-19", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.03, - "output": 0.13 + "input": 150, + "output": 600 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", - "family": "qwen", - "attachment": false, + "openai/gpt-chat-latest": { + "id": "openai/gpt-chat-latest", + "name": "OpenAI: GPT Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "structured_output": true, + "temperature": false, + "release_date": "2026-05-05", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 40960 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.29, - "output": 0.59 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "llama-3.3-70b-versatile": { - "id": "llama-3.3-70b-versatile", - "name": "Meta Llama 3.3 70B Versatile", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/gpt-5-image": { + "id": "openai/gpt-5-image", + "name": "OpenAI: GPT-5 Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2025-10-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["image", "pdf", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32678 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.59, - "output": 0.7899999999999999 + "input": 10, + "output": 10 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "OpenAI GPT-5 Mini", - "family": "gpt-mini", - "attachment": false, - "reasoning": false, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "OpenAI: GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-03-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1050000, "output": 128000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.024999999999999998 + "input": 2.5, + "output": 15 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "OpenAI GPT-5 Nano", - "family": "gpt-nano", - "attachment": false, - "reasoning": false, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "OpenAI: GPT-5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-03-17", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, @@ -64296,259 +111623,267 @@ "output": 128000 }, "cost": { - "input": 0.049999999999999996, - "output": 0.39999999999999997, - "cache_read": 0.005 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Google Gemini 3 Pro Preview", - "family": "gemini-pro", - "attachment": false, - "reasoning": true, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "OpenAI: GPT-4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "release_date": "2025-04-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1047576, + "output": 32768 }, "cost": { "input": 2, - "output": 12, - "cache_read": 0.19999999999999998 + "output": 8, + "cache_read": 0.5 } }, - "claude-3-haiku-20240307": { - "id": "claude-3-haiku-20240307", - "name": "Anthropic: Claude 3 Haiku", - "family": "claude-haiku", + "openai/gpt-4o-audio-preview": { + "id": "openai/gpt-4o-audio-preview", + "name": "OpenAI: GPT-4o Audio", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-03-07", - "last_updated": "2024-03-07", + "release_date": "2025-08-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["audio", "text"], + "output": ["audio", "text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 2.5, + "output": 10 } }, - "llama-4-maverick": { - "id": "llama-4-maverick", - "name": "Meta Llama 4 Maverick 17B 128E", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/o3-deep-research": { + "id": "openai/o3-deep-research", + "name": "OpenAI: o3 Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2024-06-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 10, + "output": 40, + "cache_read": 2.5 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Anthropic: Claude Sonnet 4.5 (20250929)", - "family": "claude-sonnet", + "openai/gpt-4-turbo-preview": { + "id": "openai/gpt-4-turbo-preview", + "name": "OpenAI: GPT-4 Turbo Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2024-01-25", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.30000000000000004, - "cache_write": 3.75 + "input": 10, + "output": 30 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Google Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": false, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "OpenAI: GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "temperature": false, + "release_date": "2025-08-07", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.3125, - "cache_write": 1.25 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "claude-4.5-opus": { - "id": "claude-4.5-opus", - "name": "Anthropic: Claude Opus 4.5", - "family": "claude-opus", - "attachment": false, - "reasoning": true, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "OpenAI: GPT-4.1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "release_date": "2025-04-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "xAI Grok 4.1 Fast Non-Reasoning", - "family": "grok", - "attachment": false, + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "OpenAI: GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-17", - "last_updated": "2025-11-17", + "release_date": "2023-09-13", + "last_updated": "2024-04-09", "modalities": { "input": ["text", "image"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.19999999999999998, - "output": 0.5, - "cache_read": 0.049999999999999996 + "input": 10, + "output": 30 } }, - "sonar-pro": { - "id": "sonar-pro", - "name": "Perplexity Sonar Pro", - "family": "sonar-pro", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "OpenAI: GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-08-07", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 400000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "mistral-large-2411": { - "id": "mistral-large-2411", - "name": "Mistral-Large", - "family": "mistral-large", - "attachment": false, - "reasoning": false, + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "OpenAI: GPT-5.4 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-24", - "last_updated": "2024-07-24", + "release_date": "2026-03-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1050000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6 + "input": 30, + "output": 180 } }, - "o3-pro": { - "id": "o3-pro", - "name": "OpenAI o3 Pro", - "family": "o-pro", - "attachment": false, + "openai/o3-mini-high": { + "id": "openai/o3-mini-high", + "name": "OpenAI: o3 Mini High", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": false, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "release_date": "2025-01-31", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["pdf", "text"], "output": ["text"] }, "open_weights": false, @@ -64557,50 +111892,48 @@ "output": 100000 }, "cost": { - "input": 20, - "output": 80 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "claude-opus-4-1": { - "id": "claude-opus-4-1", - "name": "Anthropic: Claude Opus 4.1", - "family": "claude-opus", - "attachment": false, + "openai/gpt-5.4-image-2": { + "id": "openai/gpt-5.4-image-2", + "name": "OpenAI: GPT-5.4 Image 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [], + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["image", "text", "pdf"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 272000, + "output": 128000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 8, + "output": 15, + "cache_read": 2 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "OpenAI GPT-4o-mini", - "family": "gpt-mini", + "openai/gpt-4o-search-preview": { + "id": "openai/gpt-4o-search-preview", + "name": "OpenAI: GPT-4o Search Preview", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "tool_call": false, + "release_date": "2025-03-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, @@ -64609,1228 +111942,1280 @@ "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.075 + "input": 2.5, + "output": 10 } }, - "claude-4.5-haiku": { - "id": "claude-4.5-haiku", - "name": "Anthropic: Claude 4.5 Haiku", - "family": "claude-haiku", - "attachment": false, - "reasoning": false, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "OpenAI: GPT-5.5 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-10", - "release_date": "2025-10-01", - "last_updated": "2025-10-01", + "temperature": false, + "release_date": "2026-04-24", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 1050000, + "output": 128000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.09999999999999999, - "cache_write": 1.25 + "input": 30, + "output": 180 } }, - "kimi-k2-0711": { - "id": "kimi-k2-0711", - "name": "Kimi K2 (07/11)", - "family": "kimi", - "attachment": false, + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "OpenAI: GPT-4o-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2024-07-18", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.5700000000000001, - "output": 2.3 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "o4-mini": { - "id": "o4-mini", - "name": "OpenAI o4 Mini", - "family": "o-mini", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "OpenAI: gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.275 + "input": 0.03, + "output": 0.14 } }, - "sonar-deep-research": { - "id": "sonar-deep-research", - "name": "Perplexity Sonar Deep Research", - "family": "sonar-deep-research", + "openai/gpt-4-1106-preview": { + "id": "openai/gpt-4-1106-preview", + "name": "OpenAI: GPT-4 Turbo (older v1106)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "release_date": "2023-11-06", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 127000, + "context": 128000, "output": 4096 }, "cost": { - "input": 2, - "output": 8 + "input": 10, + "output": 30 } }, - "gemma-3-12b-it": { - "id": "gemma-3-12b-it", - "name": "Google Gemma 3 12B", - "family": "gemma", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "OpenAI: GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.049999999999999996, - "output": 0.09999999999999999 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Google Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": false, + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "OpenAI: GPT-5.2-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "temperature": false, + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65535 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "cache_write": 0.3 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "deepseek-tng-r1t2-chimera": { - "id": "deepseek-tng-r1t2-chimera", - "name": "DeepSeek TNG R1T2 Chimera", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": false, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "OpenAI: GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-02", - "last_updated": "2025-07-02", + "temperature": false, + "release_date": "2025-11-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 130000, - "output": 163840 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "OpenAI: GPT-5.1 Codex Mini", - "family": "gpt-codex", - "attachment": false, - "reasoning": false, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "OpenAI: GPT-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-04-24", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1050000, "output": 128000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.024999999999999998 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "claude-sonnet-4": { - "id": "claude-sonnet-4", - "name": "Anthropic: Claude Sonnet 4", - "family": "claude-sonnet", + "thedrummer/cydonia-24b-v4.1": { + "id": "thedrummer/cydonia-24b-v4.1", + "name": "TheDrummer: Cydonia 24B V4.1", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-14", - "last_updated": "2025-05-14", + "release_date": "2025-09-27", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.30000000000000004, - "cache_write": 3.75 + "input": 0.3, + "output": 0.5 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "xAI Grok Code Fast 1", - "family": "grok", + "thedrummer/skyfall-36b-v2": { + "id": "thedrummer/skyfall-36b-v2", + "name": "TheDrummer: Skyfall 36B V2", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-25", - "last_updated": "2024-08-25", + "release_date": "2025-03-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 10000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.19999999999999998, - "output": 1.5, - "cache_read": 0.02 - } - }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "OpenAI GPT-5.1", - "family": "gpt", + "input": 0.55, + "output": 0.8 + } + }, + "thedrummer/unslopnemo-12b": { + "id": "thedrummer/unslopnemo-12b", + "name": "TheDrummer: UnslopNemo 12B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2024-11-09", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0.4, + "output": 0.4 } }, - "deepseek-reasoner": { - "id": "deepseek-reasoner", - "name": "DeepSeek Reasoner", - "family": "deepseek-thinking", + "thedrummer/rocinante-12b": { + "id": "thedrummer/rocinante-12b", + "name": "TheDrummer: Rocinante 12B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2024-09-30", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.56, - "output": 1.68, - "cache_read": 0.07 + "input": 0.17, + "output": 0.43 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "xAI: Grok 4 Fast Reasoning", - "family": "grok", - "attachment": false, - "reasoning": true, - "tool_call": true, + "bytedance/ui-tars-1.5-7b": { + "id": "bytedance/ui-tars-1.5-7b", + "name": "ByteDance: UI-TARS 7B ", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-01", - "last_updated": "2025-09-01", + "release_date": "2025-07-23", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 128000, + "output": 2048 }, "cost": { - "input": 0.19999999999999998, - "output": 0.5, - "cache_read": 0.049999999999999996 + "input": 0.1, + "output": 0.2 } }, - "o1": { - "id": "o1", - "name": "OpenAI: o1", - "family": "o", + "rekaai/reka-flash-3": { + "id": "rekaai/reka-flash-3", + "name": "Reka Flash 3", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2025-03-12", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.1, + "output": 0.2 } }, - "llama-3.1-8b-instant": { - "id": "llama-3.1-8b-instant", - "name": "Meta Llama 3.1 8B Instant", - "family": "llama", - "attachment": false, + "rekaai/reka-edge": { + "id": "rekaai/reka-edge", + "name": "Reka Edge", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "release_date": "2026-03-20", + "last_updated": "2026-04-11", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 32678 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.049999999999999996, - "output": 0.08 + "input": 0.1, + "output": 0.1 } }, - "o3-mini": { - "id": "o3-mini", - "name": "OpenAI o3 Mini", - "family": "o-mini", + "mistralai/mistral-large-2407": { + "id": "mistralai/mistral-large-2407", + "name": "Mistral Large 2407", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2023-10", - "release_date": "2023-10-01", - "last_updated": "2023-10-01", + "temperature": true, + "release_date": "2024-11-19", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 2, + "output": 6 } }, - "sonar": { - "id": "sonar", - "name": "Perplexity Sonar", - "family": "sonar", - "attachment": false, + "mistralai/mistral-small-3.2-24b-instruct": { + "id": "mistralai/mistral-small-3.2-24b-instruct", + "name": "Mistral: Mistral Small 3.2 24B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 127000, - "output": 4096 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1, - "output": 1 + "input": 0.06, + "output": 0.18, + "cache_read": 0.03 } }, - "kimi-k2-0905": { - "id": "kimi-k2-0905", - "name": "Kimi K2 (09/05)", - "family": "kimi", + "mistralai/mistral-nemo": { + "id": "mistralai/mistral-nemo", + "name": "Mistral: Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2024-07-01", + "last_updated": "2024-07-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, + "context": 131072, "output": 16384 }, "cost": { - "input": 0.5, - "output": 2, - "cache_read": 0.39999999999999997 + "input": 0.02, + "output": 0.04 } }, - "mistral-small": { - "id": "mistral-small", - "name": "Mistral Small", - "family": "mistral-small", - "attachment": false, - "reasoning": false, - "tool_call": false, + "mistralai/mistral-medium-3-5": { + "id": "mistralai/mistral-medium-3-5", + "name": "Mistral: Mistral Medium 3.5", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-02", - "release_date": "2024-02-26", - "last_updated": "2024-02-26", + "release_date": "2026-04-30", + "last_updated": "2026-05-07", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 75, - "output": 200 + "input": 1.5, + "output": 7.5 } }, - "qwen3-30b-a3b": { - "id": "qwen3-30b-a3b", - "name": "Qwen3 30B A3B", - "family": "qwen", - "attachment": false, + "mistralai/ministral-8b-2512": { + "id": "mistralai/ministral-8b-2512", + "name": "Mistral: Ministral 3 8B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "release_date": "2025-12-02", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 41000, - "output": 41000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.08, - "output": 0.29 + "input": 0.15, + "output": 0.15 } }, - "grok-4": { - "id": "grok-4", - "name": "xAI Grok 4", - "family": "grok", + "mistralai/devstral-small": { + "id": "mistralai/devstral-small", + "name": "Mistral: Devstral Small 1.1", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-09", - "last_updated": "2024-07-09", + "release_date": "2025-05-07", + "last_updated": "2025-07-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.1, + "output": 0.3 } }, - "qwen3-235b-a22b-thinking": { - "id": "qwen3-235b-a22b-thinking", - "name": "Qwen3 235B A22B Thinking", - "family": "qwen", - "attachment": false, - "reasoning": true, + "mistralai/mistral-small-3.1-24b-instruct": { + "id": "mistralai/mistral-small-3.1-24b-instruct", + "name": "Mistral: Mistral Small 3.1 24B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "attachment": true, + "reasoning": false, "tool_call": false, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2025-03-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 81920 + "context": 128000, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 2.9000000000000004 + "input": 0.35, + "output": 0.56, + "cache_read": 0.015 } }, - "qwen2.5-coder-7b-fast": { - "id": "qwen2.5-coder-7b-fast", - "name": "Qwen2.5 Coder 7B fast", - "family": "qwen", + "mistralai/mistral-saba": { + "id": "mistralai/mistral-saba", + "name": "Mistral: Saba", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-09-15", - "last_updated": "2024-09-15", + "release_date": "2025-02-17", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 8192 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.03, - "output": 0.09 + "input": 0.2, + "output": 0.6 } }, - "llama-3.1-8b-instruct-turbo": { - "id": "llama-3.1-8b-instruct-turbo", - "name": "Meta Llama 3.1 8B Instruct Turbo", - "family": "llama", + "mistralai/mistral-large": { + "id": "mistralai/mistral-large", + "name": "Mistral Large", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2024-07-24", + "last_updated": "2025-12-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 128000 + "output": 25600 }, "cost": { - "input": 0.02, - "output": 0.03 + "input": 2, + "output": 6 } }, - "qwen3-next-80b-a3b-instruct": { - "id": "qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "family": "qwen", - "attachment": false, + "mistralai/mistral-medium-3.1": { + "id": "mistralai/mistral-medium-3.1", + "name": "Mistral: Mistral Medium 3.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2025-08-12", + "last_updated": "2025-08-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 16384 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 0.4, + "output": 2 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "Zai GLM-4.6", - "family": "glm", - "attachment": false, - "reasoning": true, + "mistralai/pixtral-large-2411": { + "id": "mistralai/pixtral-large-2411", + "name": "Mistral: Pixtral Large 2411", + "description": "Mistral vision-language model for image understanding and multimodal chat", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2024-11-19", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.44999999999999996, - "output": 1.5 + "input": 2, + "output": 6 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "OpenAI: GPT-5 Codex", - "family": "gpt-codex", + "mistralai/devstral-medium": { + "id": "mistralai/devstral-medium", + "name": "Mistral: Devstral Medium", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", "attachment": false, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2025-07-10", + "last_updated": "2025-07-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0.4, + "output": 2 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "Anthropic: Claude Opus 4.1 (20250805)", - "family": "claude-opus", + "mistralai/mistral-small-24b-instruct-2501": { + "id": "mistralai/mistral-small-24b-instruct-2501", + "name": "Mistral: Mistral Small 3", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-12-29", + "last_updated": "2026-01-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 32768, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.05, + "output": 0.08 } }, - "gpt-5.1-chat-latest": { - "id": "gpt-5.1-chat-latest", - "name": "OpenAI GPT-5.1 Chat", - "family": "gpt-codex", - "attachment": false, + "mistralai/ministral-3b-2512": { + "id": "mistralai/ministral-3b-2512", + "name": "Mistral: Ministral 3 3B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2025-12-02", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["image", "text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0.1, + "output": 0.1 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Anthropic: Claude 4.5 Haiku (20251001)", - "family": "claude-haiku", - "attachment": false, - "reasoning": false, + "mistralai/mistral-small-2603": { + "id": "mistralai/mistral-small-2603", + "name": "Mistral: Mistral Small 4", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-10", - "release_date": "2025-10-01", - "last_updated": "2025-10-01", + "release_date": "2026-03-16", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.09999999999999999, - "cache_write": 1.25 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "sonar-reasoning": { - "id": "sonar-reasoning", - "name": "Perplexity Sonar Reasoning", - "family": "sonar-reasoning", + "mistralai/mistral-large-2411": { + "id": "mistralai/mistral-large-2411", + "name": "Mistral Large 2411", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "release_date": "2024-07-24", + "last_updated": "2024-11-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 127000, - "output": 4096 + "context": 131072, + "output": 26215 }, "cost": { - "input": 1, - "output": 5 + "input": 2, + "output": 6 } }, - "claude-opus-4": { - "id": "claude-opus-4", - "name": "Anthropic: Claude Opus 4", - "family": "claude-opus", + "mistralai/mistral-7b-instruct-v0.1": { + "id": "mistralai/mistral-7b-instruct-v0.1", + "name": "Mistral: Mistral 7B Instruct v0.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-14", - "last_updated": "2025-05-14", + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 2824, + "output": 565 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.11, + "output": 0.19 } }, - "llama-prompt-guard-2-86m": { - "id": "llama-prompt-guard-2-86m", - "name": "Meta Llama Prompt Guard 2 86M", - "family": "llama", - "attachment": false, + "mistralai/ministral-14b-2512": { + "id": "mistralai/ministral-14b-2512", + "name": "Mistral: Ministral 3 14B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "release_date": "2025-12-16", + "last_updated": "2025-12-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 512, - "output": 2 + "context": 262144, + "output": 52429 }, "cost": { - "input": 0.01, - "output": 0.01 + "input": 0.2, + "output": 0.2 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "OpenAI GPT-4.1 Nano", - "family": "gpt-nano", + "mistralai/devstral-2512": { + "id": "mistralai/devstral-2512", + "name": "Mistral: Devstral 2 2512", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2025-09-12", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.09999999999999999, - "output": 0.39999999999999997, - "cache_read": 0.024999999999999998 + "input": 0.4, + "output": 2, + "cache_read": 0.025 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3 Coder 30B A3B Instruct", - "family": "qwen", + "mistralai/mixtral-8x22b-instruct": { + "id": "mistralai/mixtral-8x22b-instruct", + "name": "Mistral: Mixtral 8x22B Instruct", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "release_date": "2024-04-17", + "last_updated": "2024-04-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 65536, + "output": 13108 }, "cost": { - "input": 0.09999999999999999, - "output": 0.3 + "input": 2, + "output": 6 } }, - "claude-3.5-haiku": { - "id": "claude-3.5-haiku", - "name": "Anthropic: Claude 3.5 Haiku", - "family": "claude-haiku", - "attachment": false, + "mistralai/mistral-medium-3": { + "id": "mistralai/mistral-medium-3", + "name": "Mistral: Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.7999999999999999, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0.4, + "output": 2 } }, - "grok-3-mini": { - "id": "grok-3-mini", - "name": "xAI Grok 3 Mini", - "family": "grok", + "mistralai/voxtral-small-24b-2507": { + "id": "mistralai/voxtral-small-24b-2507", + "name": "Mistral: Voxtral Small 24B 2507", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { - "input": ["text"], + "input": ["text", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 32000, + "output": 6400 }, "cost": { - "input": 0.3, - "output": 0.5, - "cache_read": 0.075 + "input": 0.1, + "output": 0.3 } }, - "o3": { - "id": "o3", - "name": "OpenAI o3", - "family": "o", - "attachment": false, + "mistralai/mistral-large-2512": { + "id": "mistralai/mistral-large-2512", + "name": "Mistral: Mistral Large 3 2512", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "temperature": true, + "release_date": "2024-11-01", + "last_updated": "2025-12-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 262144, + "output": 52429 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.5, + "output": 1.5 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "mistralai/codestral-2508": { + "id": "mistralai/codestral-2508", + "name": "Mistral: Codestral 2508", + "description": "Mistral coding model for code completion, generation, and developer workflows", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "release_date": "2025-08-01", + "last_updated": "2025-08-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 256000, + "output": 51200 }, "cost": { - "input": 0.27, - "output": 0.41 + "input": 0.3, + "output": 0.9 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "OpenAI GPT-OSS 20b", - "family": "gpt-oss", + "morph/morph-v3-fast": { + "id": "morph/morph-v3-fast", + "name": "Morph: Morph V3 Fast", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 81920, + "output": 38000 }, "cost": { - "input": 0.049999999999999996, - "output": 0.19999999999999998 + "input": 0.8, + "output": 1.2 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "OpenAI: GPT-5 Pro", - "family": "gpt-pro", + "morph/morph-v3-large": { + "id": "morph/morph-v3-large", + "name": "Morph: Morph V3 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2024-08-15", + "last_updated": "2024-08-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 262144, + "output": 131072 }, "cost": { - "input": 15, - "output": 120 + "input": 0.9, + "output": 1.9 } }, - "llama-guard-4": { - "id": "llama-guard-4", - "name": "Meta Llama Guard 4 12B", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "bytedance-seed/seed-1.6-flash": { + "id": "bytedance-seed/seed-1.6-flash", + "name": "ByteDance Seed: Seed 1.6 Flash", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2025-12-23", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 1024 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.21, - "output": 0.21 + "input": 0.075, + "output": 0.3 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "OpenAI GPT-4o", - "family": "gpt", - "attachment": false, - "reasoning": false, + "bytedance-seed/seed-1.6": { + "id": "bytedance-seed/seed-1.6", + "name": "ByteDance Seed: Seed 1.6", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-05", - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.25, + "output": 2 } }, - "qwen3-vl-235b-a22b-instruct": { - "id": "qwen3-vl-235b-a22b-instruct", - "name": "Qwen3 VL 235B A22B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "bytedance-seed/seed-2.0-mini": { + "id": "bytedance-seed/seed-2.0-mini", + "name": "ByteDance Seed: Seed-2.0-Mini", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "release_date": "2026-02-27", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.1, + "output": 0.4 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Google Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", - "attachment": false, + "bytedance-seed/seed-2.0-lite": { + "id": "bytedance-seed/seed-2.0-lite", + "name": "ByteDance Seed: Seed-2.0-Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "release_date": "2026-03-10", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65535 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.09999999999999999, - "output": 0.39999999999999997, - "cache_read": 0.024999999999999998, - "cache_write": 0.09999999999999999 + "input": 0.25, + "output": 2 } }, - "qwen3-coder": { - "id": "qwen3-coder", - "name": "Qwen3 Coder 480B A35B Instruct Turbo", - "family": "qwen", + "anthracite-org/magnum-v4-72b": { + "id": "anthracite-org/magnum-v4-72b", + "name": "Magnum v4 72B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-10-22", + "last_updated": "2026-03-15", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 16384, + "output": 2048 + }, + "cost": { + "input": 3, + "output": 5 + } + }, + "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": { + "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free", + "name": "NVIDIA: Nemotron 3 Nano Omni (free)", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-04-28", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "audio", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 16384 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.22, - "output": 0.95 + "input": 0, + "output": 0 } }, - "gpt-5": { - "id": "gpt-5", - "name": "OpenAI GPT-5", - "family": "gpt", + "nvidia/nemotron-3-nano-30b-a3b": { + "id": "nvidia/nemotron-3-nano-30b-a3b", + "name": "NVIDIA: Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2024-12", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 52429 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0.05, + "output": 0.2 } }, - "ernie-4.5-21b-a3b-thinking": { - "id": "ernie-4.5-21b-a3b-thinking", - "name": "Baidu Ernie 4.5 21B A3B Thinking", - "family": "ernie", + "nvidia/llama-3.3-nemotron-super-49b-v1.5": { + "id": "nvidia/llama-3.3-nemotron-super-49b-v1.5", + "name": "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-03", "release_date": "2025-03-16", "last_updated": "2025-03-16", "modalities": { @@ -65839,231 +113224,314 @@ }, "open_weights": false, "limit": { - "context": 128000, - "output": 8000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.07, - "output": 0.28 - } - }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "OpenAI GPT-OSS 120b", - "family": "gpt-oss", + "input": 0.1, + "output": 0.4 + } + }, + "nvidia/nemotron-nano-9b-v2": { + "id": "nvidia/nemotron-nano-9b-v2", + "name": "NVIDIA: Nemotron Nano 9B V2", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "release_date": "2025-08-18", + "last_updated": "2025-08-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 131072 + "output": 26215 }, "cost": { "input": 0.04, "output": 0.16 } }, - "gpt-5-chat-latest": { - "id": "gpt-5-chat-latest", - "name": "OpenAI GPT-5 Chat Latest", - "family": "gpt-codex", + "nvidia/nemotron-3-super-120b-a12b:free": { + "id": "nvidia/nemotron-3-super-120b-a12b:free", + "name": "NVIDIA: Nemotron 3 Super (free)", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-09", - "release_date": "2024-09-30", - "last_updated": "2024-09-30", + "temperature": true, + "release_date": "2026-03-12", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0, + "output": 0 } }, - "claude-4.5-sonnet": { - "id": "claude-4.5-sonnet", - "name": "Anthropic: Claude Sonnet 4.5", - "family": "claude-sonnet", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "NVIDIA: Nemotron 3 Super", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-03-11", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.30000000000000004, - "cache_write": 3.75 + "input": 0.1, + "output": 0.5, + "cache_read": 0.1 } }, - "deepseek-v3": { - "id": "deepseek-v3", - "name": "DeepSeek V3", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "Xiaomi: MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, "knowledge": "2024-12", - "release_date": "2024-12-26", - "last_updated": "2024-12-26", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.56, - "output": 1.68, - "cache_read": 0.07 + "input": 0.4, + "output": 2, + "cache_read": 0.08, + "tiers": [ + { + "input": 0.8, + "output": 4, + "cache_read": 0.16, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.8, + "output": 4, + "cache_read": 0.16 + } } }, - "llama-3.1-8b-instruct": { - "id": "llama-3.1-8b-instruct", - "name": "Meta Llama 3.1 8B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "xiaomi/mimo-v2-omni": { + "id": "xiaomi/mimo-v2-omni", + "name": "Xiaomi: MiMo-V2-Omni", + "description": "MiMo omni model for text, image, video, audio, and agents", + "family": "mimo", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.02, - "output": 0.049999999999999996 + "input": 0.4, + "output": 2, + "cache_read": 0.08 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "OpenAI GPT-4.1", - "family": "gpt", + "xiaomi/mimo-v2-flash": { + "id": "xiaomi/mimo-v2-flash", + "name": "Xiaomi: MiMo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2024-12-01", + "release_date": "2025-12-16", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 262144, + "output": 65536 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.09, + "output": 0.29, + "cache_read": 0.045 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "xiaomi/mimo-v2-pro": { + "id": "xiaomi/mimo-v2-pro", + "name": "Xiaomi: MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "knowledge": "2024-12", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 262144 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.48, - "output": 2 + "input": 1, + "output": 3, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "OpenAI GPT-4.1 Mini", - "family": "gpt-mini", + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "Xiaomi: MiMo V2.5 Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.39999999999999997, - "output": 1.5999999999999999, - "cache_read": 0.09999999999999999 + "input": 1, + "output": 3, + "cache_read": 0.2, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } } }, - "deepseek-v3.1-terminus": { - "id": "deepseek-v3.1-terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", + "inception/mercury-2": { + "id": "inception/mercury-2", + "name": "Inception: Mercury 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { "input": ["text"], "output": ["text"] @@ -66071,833 +113539,1045 @@ "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "output": 50000 }, "cost": { - "input": 0.27, - "output": 1, - "cache_read": 0.21600000000000003 + "input": 0.25, + "output": 0.75, + "cache_read": 0.025 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "OpenAI: GPT-5.1 Codex", - "family": "gpt-codex", - "attachment": false, + "anthropic/claude-3.5-haiku": { + "id": "anthropic/claude-3.5-haiku", + "name": "Anthropic: Claude 3.5 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "temperature": true, + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text", "image"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.12500000000000003 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "grok-3": { - "id": "grok-3", - "name": "xAI Grok 3", - "family": "grok", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Anthropic: Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "release_date": "2025-09-29", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 64000 }, "cost": { "input": 3, "output": 15, - "cache_read": 0.75 + "cache_read": 0.3, + "cache_write": 3.75 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "xAI Grok 4 Fast Non-Reasoning", - "family": "grok", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Anthropic: Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "release_date": "2025-05-22", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "audio"], + "input": ["image", "pdf", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.19999999999999998, - "output": 0.5, - "cache_read": 0.049999999999999996 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "sonar-reasoning-pro": { - "id": "sonar-reasoning-pro", - "name": "Perplexity Sonar Reasoning Pro", - "family": "sonar-reasoning", - "attachment": false, + "anthropic/claude-opus-4.6-fast": { + "id": "anthropic/claude-opus-4.6-fast", + "name": "Anthropic: Claude Opus 4.6 (Fast)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-27", - "last_updated": "2025-01-27", + "knowledge": "2025-05-31", + "release_date": "2026-04-07", + "last_updated": "2026-04-11", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 127000, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 8 - } - } - } - }, - "ollama-cloud": { - "id": "ollama-cloud", - "env": ["OLLAMA_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://ollama.com/v1", - "name": "Ollama Cloud", - "doc": "https://docs.ollama.com/cloud", - "models": { - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "minimax-m2.7", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 196608, - "output": 196608 + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 } }, - "gpt-oss:20b": { - "id": "gpt-oss:20b", - "name": "gpt-oss:20b", - "family": "gpt-oss", - "attachment": false, + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Anthropic: Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "release_date": "2025-08-05", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "kimi-k2.5", - "family": "kimi", + "anthropic/claude-opus-4.7-fast": { + "id": "anthropic/claude-opus-4.7-fast", + "name": "Anthropic: Claude Opus 4.7 (Fast)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "temperature": false, + "release_date": "2026-05-12", + "last_updated": "2026-05-16", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - } - }, - "glm-4.7": { - "id": "glm-4.7", - "name": "glm-4.7", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2025-12-22", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text"], - "output": ["text"] + "context": 1000000, + "output": 128000 }, - "open_weights": true, - "limit": { - "context": 202752, - "output": 131072 + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 } }, - "gemma4:31b": { - "id": "gemma4:31b", - "name": "gemma4:31b", - "family": "gemma", + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Anthropic: Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "knowledge": "2025-01", - "release_date": "2026-04-02", - "last_updated": "2026-04-08", + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "gpt-oss:120b": { - "id": "gpt-oss:120b", - "name": "gpt-oss:120b", - "family": "gpt-oss", - "attachment": false, + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Anthropic: Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "temperature": true, "release_date": "2025-08-05", - "last_updated": "2026-01-19", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "qwen3.5:397b": { - "id": "qwen3.5:397b", - "name": "qwen3.5:397b", - "family": "qwen", + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Anthropic: Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "release_date": "2026-02-15", - "last_updated": "2026-02-17", + "temperature": true, + "release_date": "2025-11-24", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 - } - }, - "deepseek-v3.1:671b": { - "id": "deepseek-v3.1:671b", - "name": "deepseek-v3.1:671b", - "family": "deepseek", - "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2025-08-21", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text"], - "output": ["text"] + "context": 200000, + "output": 64000 }, - "open_weights": true, - "limit": { - "context": 163840, - "output": 163840 + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "glm-5": { - "id": "glm-5", - "name": "glm-5", - "family": "glm", - "attachment": false, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Anthropic: Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 3, + "output": 15 } }, - "qwen3-vl:235b-instruct": { - "id": "qwen3-vl:235b-instruct", - "name": "qwen3-vl:235b-instruct", - "family": "qwen", + "anthropic/claude-3-haiku": { + "id": "anthropic/claude-3-haiku", + "name": "Anthropic: Claude 3 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", "attachment": true, "reasoning": false, "tool_call": true, - "release_date": "2025-09-22", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2024-03-07", + "last_updated": "2024-03-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 - } - }, - "gemma3:4b": { - "id": "gemma3:4b", - "name": "gemma3:4b", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, - "release_date": "2024-12-01", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text", "image"], - "output": ["text"] + "context": 200000, + "output": 4096 }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 131072 + "cost": { + "input": 0.25, + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "gemini-3-flash-preview", - "family": "gemini-flash", + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Anthropic: Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2026-04-08", + "temperature": true, + "release_date": "2025-05-22", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "ministral-3:14b": { - "id": "ministral-3:14b", - "name": "ministral-3:14b", - "family": "ministral", + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Anthropic: Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "release_date": "2024-12-01", - "last_updated": "2026-01-19", + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 1000000, "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "minimax-m2": { - "id": "minimax-m2", - "name": "minimax-m2", - "family": "minimax", + "tencent/hunyuan-a13b-instruct": { + "id": "tencent/hunyuan-a13b-instruct", + "name": "Tencent: Hunyuan A13B Instruct", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", "attachment": false, - "reasoning": false, - "tool_call": true, - "release_date": "2025-10-23", - "last_updated": "2026-01-19", + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-06-30", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 128000 + "context": 131072, + "output": 131072 + }, + "cost": { + "input": 0.14, + "output": 0.57 } }, - "qwen3-next:80b": { - "id": "qwen3-next:80b", - "name": "qwen3-next:80b", - "family": "qwen", + "tencent/hy3-preview": { + "id": "tencent/hy3-preview", + "name": "Tencent: Hy3 Preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], "tool_call": true, - "release_date": "2025-09-15", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2026-04-22", + "last_updated": "2026-05-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 32768 + "output": 262144 + }, + "cost": { + "input": 0.066, + "output": 0.26, + "cache_read": 0.029 } }, - "qwen3-vl:235b": { - "id": "qwen3-vl:235b", - "name": "qwen3-vl:235b", - "family": "qwen", - "attachment": true, + "deepcogito/cogito-v2.1-671b": { + "id": "deepcogito/cogito-v2.1-671b", + "name": "Deep Cogito: Cogito v2.1 671B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, - "release_date": "2025-09-22", - "last_updated": "2026-01-19", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-11-14", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 128000, "output": 32768 + }, + "cost": { + "input": 1.25, + "output": 1.25 } }, - "rnj-1:8b": { - "id": "rnj-1:8b", - "name": "rnj-1:8b", - "family": "rnj", + "cohere/command-a": { + "id": "cohere/command-a", + "name": "Cohere: Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", "attachment": false, "reasoning": false, - "tool_call": true, - "release_date": "2025-12-06", - "last_updated": "2026-01-19", + "tool_call": false, + "temperature": true, + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 4096 + "context": 256000, + "output": 8192 + }, + "cost": { + "input": 2.5, + "output": 10 } }, - "minimax-m2.1": { - "id": "minimax-m2.1", - "name": "minimax-m2.1", - "family": "minimax", + "cohere/command-r-08-2024": { + "id": "cohere/command-r-08-2024", + "name": "Cohere: Command R (08-2024)", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "release_date": "2025-12-23", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 4000 + }, + "cost": { + "input": 0.15, + "output": 0.6 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "glm-5.1", - "family": "glm", + "cohere/command-r7b-12-2024": { + "id": "cohere/command-r7b-12-2024", + "name": "Cohere: Command R7B (12-2024)", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "release_date": "2026-03-27", - "last_updated": "2026-04-07", + "temperature": true, + "release_date": "2024-12-02", + "last_updated": "2024-12-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 - } - }, - "mistral-large-3:675b": { - "id": "mistral-large-3:675b", - "name": "mistral-large-3:675b", - "family": "mistral-large", - "attachment": true, - "reasoning": false, - "tool_call": true, - "release_date": "2025-12-02", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text", "image"], - "output": ["text"] + "context": 128000, + "output": 4000 }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 262144 + "cost": { + "input": 0.0375, + "output": 0.15 } }, - "ministral-3:8b": { - "id": "ministral-3:8b", - "name": "ministral-3:8b", - "family": "ministral", - "attachment": true, + "cohere/command-r-plus-08-2024": { + "id": "cohere/command-r-plus-08-2024", + "name": "Cohere: Command R+ (08-2024)", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "attachment": false, "reasoning": false, "tool_call": true, - "release_date": "2024-12-01", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 128000 + "context": 128000, + "output": 4000 + }, + "cost": { + "input": 2.5, + "output": 10 } }, - "gemma3:12b": { - "id": "gemma3:12b", - "name": "gemma3:12b", - "family": "gemma", - "attachment": true, + "gryphe/mythomax-l2-13b": { + "id": "gryphe/mythomax-l2-13b", + "name": "MythoMax 13B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2024-12-01", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2024-04-25", + "last_updated": "2024-04-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 4096, + "output": 4096 + }, + "cost": { + "input": 0.06, + "output": 0.06 } }, - "qwen3-coder:480b": { - "id": "qwen3-coder:480b", - "name": "qwen3-coder:480b", - "family": "qwen", + "stepfun/step-3.5-flash": { + "id": "stepfun/step-3.5-flash", + "name": "StepFun: Step 3.5 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "release_date": "2025-07-22", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "output": 256000 + }, + "cost": { + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "nemotron-3-nano:30b": { - "id": "nemotron-3-nano:30b", - "name": "nemotron-3-nano:30b", - "family": "nemotron", + "prime-intellect/intellect-3": { + "id": "prime-intellect/intellect-3", + "name": "Prime Intellect: INTELLECT-3", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2025-12-15", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2025-11-26", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, + "context": 131072, "output": 131072 + }, + "cost": { + "input": 0.2, + "output": 1.1 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "deepseek-v4-flash", - "family": "deepseek-flash", + "nex-agi/deepseek-v3.1-nex-n1": { + "id": "nex-agi/deepseek-v3.1-nex-n1", + "name": "Nex AGI: DeepSeek V3.1 Nex N1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-11-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 1048576 + "context": 131072, + "output": 163840 + }, + "cost": { + "input": 0.27, + "output": 1 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "glm-4.6", - "family": "glm", + "undi95/remm-slerp-l2-13b": { + "id": "undi95/remm-slerp-l2-13b", + "name": "ReMM SLERP 13B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2025-09-29", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2023-07-22", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 6144, + "output": 4096 + }, + "cost": { + "input": 0.45, + "output": 0.65 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "kimi-k2.6", - "family": "kimi", + "~openai/gpt-mini-latest": { + "id": "~openai/gpt-mini-latest", + "name": "OpenAI: GPT Mini Latest", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "temperature": false, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "output": 128000 + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "ministral-3:3b": { - "id": "ministral-3:3b", - "name": "ministral-3:3b", - "family": "ministral", + "~openai/gpt-latest": { + "id": "~openai/gpt-latest", + "name": "OpenAI: GPT Latest", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "release_date": "2024-10-22", - "last_updated": "2026-01-19", + "temperature": false, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 1050000, "output": 128000 + }, + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "gemma3:27b": { - "id": "gemma3:27b", - "name": "gemma3:27b", - "family": "gemma", + "~moonshotai/kimi-latest": { + "id": "~moonshotai/kimi-latest", + "name": "MoonshotAI: Kimi Latest", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, - "reasoning": false, - "tool_call": false, - "release_date": "2025-07-27", - "last_updated": "2026-01-19", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 262142, + "output": 262142 + }, + "cost": { + "input": 0.74, + "output": 3.49, + "cache_read": 0.14 } }, - "devstral-2:123b": { - "id": "devstral-2:123b", - "name": "devstral-2:123b", - "family": "devstral", + "relace/relace-search": { + "id": "relace/relace-search", + "name": "Relace: Relace Search", + "description": "Tool-capable chat model for instruction following and agentic application workflows", "attachment": false, "reasoning": false, "tool_call": true, + "temperature": true, "release_date": "2025-12-09", - "last_updated": "2026-01-19", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - } - }, - "cogito-2.1:671b": { - "id": "cogito-2.1:671b", - "name": "cogito-2.1:671b", - "family": "cogito", - "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2025-11-19", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text"], - "output": ["text"] + "context": 256000, + "output": 128000 }, - "open_weights": true, - "limit": { - "context": 163840, - "output": 32000 + "cost": { + "input": 1, + "output": 3 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "qwen3-coder-next", - "family": "qwen", + "relace/relace-apply-3": { + "id": "relace/relace-apply-3", + "name": "Relace: Relace Apply 3", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "release_date": "2026-02-02", - "last_updated": "2026-02-08", + "tool_call": false, + "release_date": "2025-09-26", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "output": 128000 + }, + "cost": { + "input": 0.85, + "output": 1.25 } }, - "nemotron-3-super": { - "id": "nemotron-3-super", - "name": "nemotron-3-super", - "family": "nemotron", + "ai21/jamba-large-1.7": { + "id": "ai21/jamba-large-1.7", + "name": "AI21: Jamba Large 1.7", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-12", + "temperature": true, + "release_date": "2025-08-09", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "output": 4096 + }, + "cost": { + "input": 2, + "output": 8 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "deepseek-v4-pro", - "family": "deepseek-thinking", + "arcee-ai/coder-large": { + "id": "arcee-ai/coder-large", + "name": "Arcee AI: Coder Large", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-05-06", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 1048576 + "context": 32768, + "output": 32768 + }, + "cost": { + "input": 0.5, + "output": 0.8 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "minimax-m2.5", - "family": "minimax", + "arcee-ai/virtuoso-large": { + "id": "arcee-ai/virtuoso-large", + "name": "Arcee AI: Virtuoso Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "knowledge": "2025-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "temperature": true, + "release_date": "2025-05-06", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 64000 + }, + "cost": { + "input": 0.75, + "output": 1.2 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "deepseek-v3.2", - "family": "deepseek", - "attachment": false, - "reasoning": true, - "tool_call": true, - "release_date": "2025-06-15", - "last_updated": "2026-01-19", + "arcee-ai/spotlight": { + "id": "arcee-ai/spotlight", + "name": "Arcee AI: Spotlight", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-05-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 131072, + "output": 65537 + }, + "cost": { + "input": 0.18, + "output": 0.18 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "kimi-k2-thinking", - "family": "kimi-thinking", + "arcee-ai/trinity-large-thinking": { + "id": "arcee-ai/trinity-large-thinking", + "name": "Arcee AI: Trinity Large Thinking", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2026-01-19", + "temperature": true, + "release_date": "2026-04-01", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], "output": ["text"] @@ -66906,223 +114586,194 @@ "limit": { "context": 262144, "output": 262144 - } - }, - "devstral-small-2:24b": { - "id": "devstral-small-2:24b", - "name": "devstral-small-2:24b", - "family": "devstral", - "attachment": true, - "reasoning": false, - "tool_call": true, - "release_date": "2025-12-09", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text", "image"], - "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 262144 + "cost": { + "input": 0.22, + "output": 0.85 } }, - "kimi-k2:1t": { - "id": "kimi-k2:1t", - "name": "kimi-k2:1t", - "family": "kimi", + "arcee-ai/maestro-reasoning": { + "id": "arcee-ai/maestro-reasoning", + "name": "Arcee AI: Maestro Reasoning", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "knowledge": "2024-10", - "release_date": "2025-07-11", - "last_updated": "2026-01-19", + "tool_call": false, + "temperature": true, + "release_date": "2025-05-06", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 131072, + "output": 32000 + }, + "cost": { + "input": 0.9, + "output": 3.3 } - } - } - }, - "zai-coding-plan": { - "id": "zai-coding-plan", - "env": ["ZHIPU_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.z.ai/api/coding/paas/v4", - "name": "Z.AI Coding Plan", - "doc": "https://docs.z.ai/devpack/overview", - "models": { - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + }, + "arcee-ai/trinity-mini": { + "id": "arcee-ai/trinity-mini", + "name": "Arcee AI: Trinity Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2025-12", + "last_updated": "2026-01-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 131072, "output": 131072 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.045, + "output": 0.15 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "mancer/weaver": { + "id": "mancer/weaver", + "name": "Mancer: Weaver (alpha)", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "release_date": "2023-08-02", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 8000, + "output": 2000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.75, + "output": 1 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM-4.5-Air", - "family": "glm-air", - "attachment": false, + "perplexity/sonar-reasoning-pro": { + "id": "perplexity/sonar-reasoning-pro", + "name": "Perplexity: Sonar Reasoning Pro", + "description": "Web-grounded reasoning model for multi-step research and cited answers", + "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 128000, + "output": 25600 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 2, + "output": 8 } }, - "glm-5-turbo": { - "id": "glm-5-turbo", - "name": "GLM-5-Turbo", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "perplexity/sonar": { + "id": "perplexity/sonar", + "name": "Perplexity: Sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2024-01-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "structured_output": true, + "open_weights": false, + "limit": { + "context": 127072, + "output": 25415 + }, + "cost": { + "input": 1, + "output": 1 + } + }, + "perplexity/sonar-pro": { + "id": "perplexity/sonar-pro", + "name": "Perplexity: Sonar Pro", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "attachment": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 131072 + "output": 8000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 3, + "output": 15 } }, - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM-5V-Turbo", - "family": "glm", + "perplexity/sonar-pro-search": { + "id": "perplexity/sonar-pro-search", + "name": "Perplexity: Sonar Pro Search", + "description": "Advanced Sonar search model for deeper research and cited synthesis", "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "release_date": "2025-10-31", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "video", "pdf"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 131072 + "output": 8000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 3, + "output": 15 } - } - } - }, - "amazon-bedrock": { - "id": "amazon-bedrock", - "env": ["AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY", "AWS_REGION", "AWS_BEARER_TOKEN_BEDROCK"], - "npm": "@ai-sdk/amazon-bedrock", - "name": "Amazon Bedrock", - "doc": "https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html", - "models": { - "openai.gpt-oss-safeguard-120b": { - "id": "openai.gpt-oss-safeguard-120b", - "name": "GPT OSS Safeguard 120B", - "family": "gpt-oss", + }, + "perplexity/sonar-deep-research": { + "id": "perplexity/sonar-deep-research", + "name": "Perplexity: Sonar Deep Research", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", "modalities": { "input": ["text"], "output": ["text"] @@ -67130,4156 +114781,4322 @@ "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 25600 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 2, + "output": 8 } }, - "nvidia.nemotron-nano-3-30b": { - "id": "nvidia.nemotron-nano-3-30b", - "name": "NVIDIA Nemotron Nano 3 30B", - "family": "nemotron", + "switchpoint/router": { + "id": "switchpoint/router", + "name": "Switchpoint Router", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2025-07-12", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.06, - "output": 0.24 + "input": 0.85, + "output": 3.4 } }, - "nvidia.nemotron-super-3-120b": { - "id": "nvidia.nemotron-super-3-120b", - "name": "NVIDIA Nemotron 3 Super 120B A12B", - "family": "nemotron", + "openrouter/bodybuilder": { + "id": "openrouter/bodybuilder", + "name": "Body Builder (beta)", + "description": "Preview model for early access evaluation, prototyping, and compatibility testing", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "reasoning": false, + "tool_call": false, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 128000, + "output": 32768 }, + "status": "beta", "cost": { - "input": 0.15, - "output": 0.65 + "input": 0, + "output": 0 } }, - "writer.palmyra-x5-v1:0": { - "id": "writer.palmyra-x5-v1:0", - "name": "Palmyra X5", - "family": "palmyra", - "attachment": false, + "openrouter/free": { + "id": "openrouter/free", + "name": "Free Models Router", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "release_date": "2026-02-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1040000, - "output": 8192 + "context": 200000, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 6 + "input": 0, + "output": 0 } }, - "mistral.ministral-3-8b-instruct": { - "id": "mistral.ministral-3-8b-instruct", - "name": "Ministral 3 8B", - "family": "ministral", + "openrouter/owl-alpha": { + "id": "openrouter/owl-alpha", + "name": "Owl Alpha", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2026-04-28", + "last_updated": "2026-04-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1048756, + "output": 262144 }, + "status": "alpha", "cost": { - "input": 0.15, - "output": 0.15 + "input": 0, + "output": 0 } }, - "au.anthropic.claude-opus-4-6-v1": { - "id": "au.anthropic.claude-opus-4-6-v1", - "name": "AU Anthropic Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "openrouter/pareto-code": { + "id": "openrouter/pareto-code", + "name": "Pareto Code Router", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": false, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 200000, + "output": 65536 }, "cost": { - "input": 16.5, - "output": 82.5, - "cache_read": 1.65, - "cache_write": 20.625 + "input": 0, + "output": 0 } }, - "mistral.ministral-3-3b-instruct": { - "id": "mistral.ministral-3-3b-instruct", - "name": "Ministral 3 3B", - "family": "ministral", - "attachment": false, - "reasoning": false, + "openrouter/auto": { + "id": "openrouter/auto", + "name": "Auto Router", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["audio", "image", "pdf", "text", "video"], + "output": ["image", "text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 2000000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0, + "output": 0 } }, - "anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", + "qwen/qwen3.5-plus-20260420": { + "id": "qwen/qwen3.5-plus-20260420", + "name": "Qwen: Qwen3.5 Plus 2026-04-20", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.4, + "output": 2.4 } }, - "mistral.devstral-2-123b": { - "id": "mistral.devstral-2-123b", - "name": "Devstral 2 123B", - "family": "devstral", - "attachment": false, - "reasoning": false, + "qwen/qwen3-vl-235b-a22b-thinking": { + "id": "qwen/qwen3-vl-235b-a22b-thinking", + "name": "Qwen: Qwen3 VL 235B A22B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2025-09-24", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.26, + "output": 2.6 } }, - "global.anthropic.claude-opus-4-5-20251101-v1:0": { - "id": "global.anthropic.claude-opus-4-5-20251101-v1:0", - "name": "Claude Opus 4.5 (Global)", - "family": "claude-opus", + "qwen/qwen3-vl-30b-a3b-thinking": { + "id": "qwen/qwen3-vl-30b-a3b-thinking", + "name": "Qwen: Qwen3 VL 30B A3B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "release_date": "2025-10-11", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.13, + "output": 1.56 } }, - "mistral.voxtral-small-24b-2507": { - "id": "mistral.voxtral-small-24b-2507", - "name": "Voxtral Small 24B 2507", - "family": "mistral", - "attachment": true, + "qwen/qwen3-coder-plus": { + "id": "qwen/qwen3-coder-plus", + "name": "Qwen: Qwen3 Coder Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, "release_date": "2025-07-01", - "last_updated": "2025-07-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.35 + "input": 0.65, + "output": 3.25, + "cache_read": 0.2 } }, - "google.gemma-3-12b-it": { - "id": "google.gemma-3-12b-it", - "name": "Google Gemma 3 12B", - "family": "gemma", + "qwen/qwen-plus": { + "id": "qwen/qwen-plus", + "name": "Qwen: Qwen-Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": true, + "tool_call": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2024-01-25", + "last_updated": "2025-09-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.049999999999999996, - "output": 0.09999999999999999 + "input": 0.4, + "output": 1.2, + "cache_read": 0.08 } }, - "amazon.nova-pro-v1:0": { - "id": "amazon.nova-pro-v1:0", - "name": "Nova Pro", - "family": "nova-pro", - "attachment": true, + "qwen/qwen3-coder-30b-a3b-instruct": { + "id": "qwen/qwen3-coder-30b-a3b-instruct", + "name": "Qwen: Qwen3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "release_date": "2025-07-31", + "last_updated": "2025-07-31", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 300000, - "output": 8192 + "context": 160000, + "output": 32768 }, "cost": { - "input": 0.8, - "output": 3.2, - "cache_read": 0.2 + "input": 0.07, + "output": 0.27 } }, - "anthropic.claude-haiku-4-5-20251001-v1:0": { - "id": "anthropic.claude-haiku-4-5-20251001-v1:0", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, + "qwen/qwen3-32b": { + "id": "qwen/qwen3-32b", + "name": "Qwen: Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2024-12-01", + "last_updated": "2026-02-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 40960, + "output": 40960 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.08, + "output": 0.24, + "cache_read": 0.04 } }, - "minimax.minimax-m2": { - "id": "minimax.minimax-m2", - "name": "MiniMax M2", - "family": "minimax", + "qwen/qwen3-next-80b-a3b-instruct": { + "id": "qwen/qwen3-next-80b-a3b-instruct", + "name": "Qwen: Qwen3 Next 80B A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "release_date": "2025-09-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204608, - "output": 128000 + "context": 131072, + "output": 52429 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.09, + "output": 1.1 } }, - "global.anthropic.claude-opus-4-7": { - "id": "global.anthropic.claude-opus-4-7", - "name": "Claude Opus 4.7 (Global)", - "family": "claude-opus", + "qwen/qwen3-vl-8b-instruct": { + "id": "qwen/qwen3-vl-8b-instruct", + "name": "Qwen: Qwen3 VL 8B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "release_date": "2025-10-15", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.08, + "output": 0.5 } }, - "mistral.pixtral-large-2502-v1:0": { - "id": "mistral.pixtral-large-2502-v1:0", - "name": "Pixtral Large (25.02)", - "family": "mistral", - "attachment": false, - "reasoning": false, - "tool_call": true, + "qwen/qwen3.6-35b-a3b": { + "id": "qwen/qwen3.6-35b-a3b", + "name": "Qwen: Qwen3.6 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, "temperature": true, - "release_date": "2025-04-08", - "last_updated": "2025-04-08", + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 65536 }, "cost": { - "input": 2, - "output": 6 + "input": 0.1612, + "output": 0.96525, + "cache_read": 0.1612 } }, - "meta.llama4-maverick-17b-instruct-v1:0": { - "id": "meta.llama4-maverick-17b-instruct-v1:0", - "name": "Llama 4 Maverick 17B Instruct", - "family": "llama", - "attachment": true, - "reasoning": false, + "qwen/qwen3.7-max": { + "id": "qwen/qwen3.7-max", + "name": "Qwen: Qwen3.7 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-08-26", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 16384 + "output": 65536 }, "cost": { - "input": 0.24, - "output": 0.97 + "input": 1.625, + "output": 4.875, + "cache_read": 0.1625, + "cache_write": 2.03125 } }, - "us.anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "us.anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5 (US)", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen: Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2025-09-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.2, + "output": 6, + "cache_read": 0.24 } }, - "us.anthropic.claude-haiku-4-5-20251001-v1:0": { - "id": "us.anthropic.claude-haiku-4-5-20251001-v1:0", - "name": "Claude Haiku 4.5 (US)", - "family": "claude-haiku", - "attachment": true, + "qwen/qwen3-8b": { + "id": "qwen/qwen3-8b", + "name": "Qwen: Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2025-04", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 40960, + "output": 8192 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.05, + "output": 0.4, + "cache_read": 0.05 } }, - "amazon.nova-micro-v1:0": { - "id": "amazon.nova-micro-v1:0", - "name": "Nova Micro", - "family": "nova-micro", + "qwen/qwen-plus-2025-07-28": { + "id": "qwen/qwen-plus-2025-07-28", + "name": "Qwen: Qwen Plus 0728", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "release_date": "2025-09-09", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.035, - "output": 0.14, - "cache_read": 0.00875 + "input": 0.26, + "output": 0.78 } }, - "global.anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "global.anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5 (Global)", - "family": "claude-sonnet", + "qwen/qwen3.5-flash-02-23": { + "id": "qwen/qwen3.5-flash-02-23", + "name": "Qwen: Qwen3.5-Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.1, + "output": 0.4 } }, - "openai.gpt-oss-20b-1:0": { - "id": "openai.gpt-oss-20b-1:0", - "name": "gpt-oss-20b", - "family": "gpt-oss", + "qwen/qwen3-30b-a3b-instruct-2507": { + "id": "qwen/qwen3-30b-a3b-instruct-2507", + "name": "Qwen: Qwen3 30B A3B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-07-29", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.07, - "output": 0.3 + "input": 0.09, + "output": 0.3, + "cache_read": 0.04 } }, - "zai.glm-5": { - "id": "zai.glm-5", - "name": "GLM-5", - "family": "glm", + "qwen/qwen-2.5-coder-32b-instruct": { + "id": "qwen/qwen-2.5-coder-32b-instruct", + "name": "Qwen2.5 Coder 32B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2024-11-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 101376 + "context": 32768, + "output": 8192 }, "cost": { - "input": 1, - "output": 3.2 + "input": 0.2, + "output": 0.2, + "cache_read": 0.015 } }, - "qwen.qwen3-32b-v1:0": { - "id": "qwen.qwen3-32b-v1:0", - "name": "Qwen3 32B (dense)", - "family": "qwen", + "qwen/qwen3-next-80b-a3b-thinking": { + "id": "qwen/qwen3-next-80b-a3b-thinking", + "name": "Qwen: Qwen3 Next 80B A3B Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "release_date": "2025-09-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.0975, + "output": 0.78 } }, - "deepseek.v3.2": { - "id": "deepseek.v3.2", - "name": "DeepSeek-V3.2", - "family": "deepseek", + "qwen/qwen3-235b-a22b-thinking-2507": { + "id": "qwen/qwen3-235b-a22b-thinking-2507", + "name": "Qwen: Qwen3 235B A22B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "release_date": "2025-07-25", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 81920 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.62, - "output": 1.85 + "input": 0.11, + "output": 0.6 } }, - "eu.anthropic.claude-haiku-4-5-20251001-v1:0": { - "id": "eu.anthropic.claude-haiku-4-5-20251001-v1:0", - "name": "Claude Haiku 4.5 (EU)", - "family": "claude-haiku", + "qwen/qwen3-vl-32b-instruct": { + "id": "qwen/qwen3-vl-32b-instruct", + "name": "Qwen: Qwen3 VL 32B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2025-10-21", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.104, + "output": 0.416 } }, - "zai.glm-4.7-flash": { - "id": "zai.glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + "qwen/qwen3-coder": { + "id": "qwen/qwen3-coder", + "name": "Qwen: Qwen3 Coder 480B A35B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 262144, + "output": 52429 }, "cost": { - "input": 0.07, - "output": 0.4 + "input": 0.22, + "output": 1, + "cache_read": 0.022 } }, - "us.anthropic.claude-opus-4-7": { - "id": "us.anthropic.claude-opus-4-7", - "name": "Claude Opus 4.7 (US)", - "family": "claude-opus", + "qwen/qwen3.6-flash": { + "id": "qwen/qwen3.6-flash", + "name": "Qwen: Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } - }, - "amazon.nova-2-lite-v1:0": { - "id": "amazon.nova-2-lite-v1:0", - "name": "Nova 2 Lite", - "family": "nova", - "attachment": false, - "reasoning": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.33, - "output": 2.75 + "input": 0.25, + "output": 1.5, + "cache_write": 0.3125 } }, - "anthropic.claude-opus-4-5-20251101-v1:0": { - "id": "anthropic.claude-opus-4-5-20251101-v1:0", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "qwen/qwen3.5-plus-02-15": { + "id": "qwen/qwen3.5-plus-02-15", + "name": "Qwen: Qwen3.5 Plus 2026-02-15", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "release_date": "2026-02-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.26, + "output": 1.56 } }, - "qwen.qwen3-coder-480b-a35b-v1:0": { - "id": "qwen.qwen3-coder-480b-a35b-v1:0", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", + "qwen/qwen-2.5-7b-instruct": { + "id": "qwen/qwen-2.5-7b-instruct", + "name": "Qwen: Qwen2.5 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "release_date": "2024-09", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 32768, + "output": 6554 }, "cost": { - "input": 0.22, - "output": 1.8 + "input": 0.04, + "output": 0.1 } }, - "amazon.nova-lite-v1:0": { - "id": "amazon.nova-lite-v1:0", - "name": "Nova Lite", - "family": "nova-lite", + "qwen/qwen3-vl-8b-thinking": { + "id": "qwen/qwen3-vl-8b-thinking", + "name": "Qwen: Qwen3 VL 8B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "release_date": "2025-10-15", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.06, - "output": 0.24, - "cache_read": 0.015 + "input": 0.117, + "output": 1.365 } }, - "meta.llama3-1-8b-instruct-v1:0": { - "id": "meta.llama3-1-8b-instruct-v1:0", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "qwen/qwen3-max-thinking": { + "id": "qwen/qwen3-max-thinking", + "name": "Qwen: Qwen3 Max Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-01-23", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.22, - "output": 0.22 + "input": 0.78, + "output": 3.9 } }, - "anthropic.claude-opus-4-7": { - "id": "anthropic.claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, + "qwen/qwen3-30b-a3b-thinking-2507": { + "id": "qwen/qwen3-30b-a3b-thinking-2507", + "name": "Qwen: Qwen3 30B A3B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 32768, + "output": 6554 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.051, + "output": 0.34 } }, - "google.gemma-3-27b-it": { - "id": "google.gemma-3-27b-it", - "name": "Google Gemma 3 27B Instruct", - "family": "gemma", + "qwen/qwen2.5-vl-72b-instruct": { + "id": "qwen/qwen2.5-vl-72b-instruct", + "name": "Qwen: Qwen2.5 VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-27", - "last_updated": "2025-07-27", + "release_date": "2025-02-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 8192 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.12, - "output": 0.2 + "input": 0.8, + "output": 0.8, + "cache_read": 0.075 } }, - "global.anthropic.claude-haiku-4-5-20251001-v1:0": { - "id": "global.anthropic.claude-haiku-4-5-20251001-v1:0", - "name": "Claude Haiku 4.5 (Global)", - "family": "claude-haiku", + "qwen/qwen3.5-27b": { + "id": "qwen/qwen3.5-27b", + "name": "Qwen: Qwen3.5-27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.195, + "output": 1.56 } }, - "google.gemma-3-4b-it": { - "id": "google.gemma-3-4b-it", - "name": "Gemma 3 4B IT", - "family": "gemma", + "qwen/qwen3-235b-a22b": { + "id": "qwen/qwen3-235b-a22b", + "name": "Qwen: Qwen3 235B A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 38912 + } + ], "tool_call": true, "temperature": true, "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.04, - "output": 0.08 + "input": 0.455, + "output": 1.82, + "cache_read": 0.15 } }, - "meta.llama4-scout-17b-instruct-v1:0": { - "id": "meta.llama4-scout-17b-instruct-v1:0", - "name": "Llama 4 Scout 17B Instruct", - "family": "llama", - "attachment": true, + "qwen/qwen-2.5-72b-instruct": { + "id": "qwen/qwen-2.5-72b-instruct", + "name": "Qwen2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2024-09", + "last_updated": "2026-01-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 3500000, + "context": 32768, "output": 16384 }, "cost": { - "input": 0.17, - "output": 0.66 + "input": 0.12, + "output": 0.39 } }, - "deepseek.v3-v1:0": { - "id": "deepseek.v3-v1:0", - "name": "DeepSeek-V3.1", - "family": "deepseek", + "qwen/qwen-plus-2025-07-28:thinking": { + "id": "qwen/qwen-plus-2025-07-28:thinking", + "name": "Qwen: Qwen Plus 0728 (thinking)", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "release_date": "2025-09-09", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 81920 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.58, - "output": 1.68 + "input": 0.26, + "output": 0.78 } }, - "mistral.magistral-small-2509": { - "id": "mistral.magistral-small-2509", - "name": "Magistral Small 1.2", - "family": "magistral", + "qwen/qwen3-coder-next": { + "id": "qwen/qwen3-coder-next", + "name": "Qwen: Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2026-02-02", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 40000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.12, + "output": 0.75, + "cache_read": 0.035 } }, - "qwen.qwen3-next-80b-a3b": { - "id": "qwen.qwen3-next-80b-a3b", - "name": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "qwen/qwen3.6-27b": { + "id": "qwen/qwen3.6-27b", + "name": "Qwen: Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 0.325, + "output": 3.25 } }, - "zai.glm-4.7": { - "id": "zai.glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "qwen/qwen3.5-35b-a3b": { + "id": "qwen/qwen3.5-35b-a3b", + "name": "Qwen: Qwen3.5-35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.1625, + "output": 1.3 } }, - "moonshot.kimi-k2-thinking": { - "id": "moonshot.kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "qwen/qwen3.5-9b": { + "id": "qwen/qwen3.5-9b", + "name": "Qwen: Qwen3.5-9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2026-03-10", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { "context": 256000, - "output": 256000 + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.05, + "output": 0.15 } }, - "us.anthropic.claude-opus-4-5-20251101-v1:0": { - "id": "us.anthropic.claude-opus-4-5-20251101-v1:0", - "name": "Claude Opus 4.5 (US)", - "family": "claude-opus", + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen: Qwen3.5 397B A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "release_date": "2026-02-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.39, + "output": 2.34 } }, - "mistral.ministral-3-14b-instruct": { - "id": "mistral.ministral-3-14b-instruct", - "name": "Ministral 14B 3.0", - "family": "ministral", - "attachment": false, + "qwen/qwen3-vl-30b-a3b-instruct": { + "id": "qwen/qwen3-vl-30b-a3b-instruct", + "name": "Qwen: Qwen3 VL 30B A3B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-10-05", + "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.13, + "output": 0.52 } }, - "deepseek.r1-v1:0": { - "id": "deepseek.r1-v1:0", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", + "qwen/qwen3-235b-a22b-2507": { + "id": "qwen/qwen3-235b-a22b-2507", + "name": "Qwen: Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-29", + "release_date": "2025-04", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 262144, + "output": 52429 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.071, + "output": 0.1 } }, - "mistral.voxtral-mini-3b-2507": { - "id": "mistral.voxtral-mini-3b-2507", - "name": "Voxtral Mini 3B 2507", - "family": "mistral", + "qwen/qwen3-coder-flash": { + "id": "qwen/qwen3-coder-flash", + "name": "Qwen: Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-07-23", + "last_updated": "2026-03-15", "modalities": { - "input": ["audio", "text"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 0.195, + "output": 0.975, + "cache_read": 0.06 } }, - "openai.gpt-oss-120b-1:0": { - "id": "openai.gpt-oss-120b-1:0", - "name": "gpt-oss-120b", - "family": "gpt-oss", + "qwen/qwen3-14b": { + "id": "qwen/qwen3-14b", + "name": "Qwen: Qwen3 14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-04", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 40960, + "output": 40960 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.06, + "output": 0.24, + "cache_read": 0.025 } }, - "nvidia.nemotron-nano-12b-v2": { - "id": "nvidia.nemotron-nano-12b-v2", - "name": "NVIDIA Nemotron Nano 12B v2 VL BF16", - "family": "nemotron", - "attachment": false, + "qwen/qwen3-vl-235b-a22b-instruct": { + "id": "qwen/qwen3-vl-235b-a22b-instruct", + "name": "Qwen: Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2025-09-23", + "last_updated": "2026-01-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 52429 }, "cost": { "input": 0.2, - "output": 0.6 - } - }, - "eu.anthropic.claude-opus-4-7": { - "id": "eu.anthropic.claude-opus-4-7", - "name": "Claude Opus 4.7 (EU)", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 128000 - }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "output": 0.88, + "cache_read": 0.11 } }, - "minimax.minimax-m2.5": { - "id": "minimax.minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", + "qwen/qwen3-30b-a3b": { + "id": "qwen/qwen3-30b-a3b", + "name": "Qwen: Qwen3 30B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2025-04", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 98304 + "context": 40960, + "output": 40960 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.08, + "output": 0.28, + "cache_read": 0.03 } }, - "meta.llama3-3-70b-instruct-v1:0": { - "id": "meta.llama3-3-70b-instruct-v1:0", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + "qwen/qwen3.6-max-preview": { + "id": "qwen/qwen3.6-max-preview", + "name": "Qwen: Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 131072 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-04-27", + "last_updated": "2026-05-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.72, - "output": 0.72 + "input": 1.04, + "output": 6.24, + "cache_write": 1.3 } }, - "meta.llama3-1-70b-instruct-v1:0": { - "id": "meta.llama3-1-70b-instruct-v1:0", - "name": "Llama 3.1 70B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen: Qwen3.5-122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-02-26", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.72, - "output": 0.72 + "input": 0.26, + "output": 2.08 } }, - "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "eu.anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5 (EU)", - "family": "claude-sonnet", + "qwen/qwen3.6-plus": { + "id": "qwen/qwen3.6-plus", + "name": "Qwen: Qwen3.6 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2025-08-26", + "last_updated": "2026-04-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.325, + "output": 1.95, + "cache_read": 0.0325, + "cache_write": 0.40625 } }, - "eu.anthropic.claude-opus-4-5-20251101-v1:0": { - "id": "eu.anthropic.claude-opus-4-5-20251101-v1:0", - "name": "Claude Opus 4.5 (EU)", - "family": "claude-opus", + "amazon/nova-lite-v1": { + "id": "amazon/nova-lite-v1", + "name": "Amazon: Nova Lite 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "release_date": "2024-12-06", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } - }, - "moonshotai.kimi-k2.5": { - "id": "moonshotai.kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "release_date": "2026-02-06", - "last_updated": "2026-02-06", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 256000, - "output": 256000 + "context": 300000, + "output": 5120 }, "cost": { - "input": 0.6, - "output": 3 + "input": 0.06, + "output": 0.24 } }, - "au.anthropic.claude-sonnet-4-6": { - "id": "au.anthropic.claude-sonnet-4-6", - "name": "AU Anthropic Claude Sonnet 4.6", - "family": "claude-sonnet", + "amazon/nova-premier-v1": { + "id": "amazon/nova-premier-v1", + "name": "Amazon: Nova Premier 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2025-11-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 128000 + "output": 32000 }, "cost": { - "input": 3.3, - "output": 16.5, - "cache_read": 0.33, - "cache_write": 4.125 + "input": 2.5, + "output": 12.5 } }, - "openai.gpt-oss-safeguard-20b": { - "id": "openai.gpt-oss-safeguard-20b", - "name": "GPT OSS Safeguard 20B", - "family": "gpt-oss", - "attachment": false, + "amazon/nova-pro-v1": { + "id": "amazon/nova-pro-v1", + "name": "Amazon: Nova Pro 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 300000, + "output": 5120 }, "cost": { - "input": 0.07, - "output": 0.2 + "input": 0.8, + "output": 3.2 } }, - "qwen.qwen3-coder-30b-a3b-v1:0": { - "id": "qwen.qwen3-coder-30b-a3b-v1:0", - "name": "Qwen3 Coder 30B A3B Instruct", - "family": "qwen", + "amazon/nova-micro-v1": { + "id": "amazon/nova-micro-v1", + "name": "Amazon: Nova Micro 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "release_date": "2024-12-06", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 128000, + "output": 5120 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.035, + "output": 0.14 } }, - "minimax.minimax-m2.1": { - "id": "minimax.minimax-m2.1", - "name": "MiniMax M2.1", - "family": "minimax", - "attachment": false, + "amazon/nova-2-lite-v1": { + "id": "amazon/nova-2-lite-v1", + "name": "Amazon: Nova 2 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2024-12-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1000000, + "output": 65535 }, "cost": { "input": 0.3, - "output": 1.2 + "output": 2.5 } }, - "qwen.qwen3-vl-235b-a22b": { - "id": "qwen.qwen3-vl-235b-a22b", - "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", - "family": "qwen", - "attachment": true, + "aion-labs/aion-rp-llama-3.1-8b": { + "id": "aion-labs/aion-rp-llama-3.1-8b", + "name": "AionLabs: Aion-RP 1.0 (8B)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "release_date": "2025-02-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.8, + "output": 1.6 } }, - "qwen.qwen3-coder-next": { - "id": "qwen.qwen3-coder-next", - "name": "Qwen3 Coder Next", - "family": "qwen", + "aion-labs/aion-1.0-mini": { + "id": "aion-labs/aion-1.0-mini", + "name": "AionLabs: Aion-1.0-Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "release_date": "2025-02-05", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, - "output": 65536 + "output": 32768 }, "cost": { - "input": 0.22, - "output": 1.8 + "input": 0.7, + "output": 1.4 } }, - "nvidia.nemotron-nano-9b-v2": { - "id": "nvidia.nemotron-nano-9b-v2", - "name": "NVIDIA Nemotron Nano 9B v2", - "family": "nemotron", + "aion-labs/aion-2.0": { + "id": "aion-labs/aion-2.0", + "name": "AionLabs: Aion-2.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "release_date": "2026-02-24", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.06, - "output": 0.23 + "input": 0.8, + "output": 1.6 } }, - "mistral.mistral-large-3-675b-instruct": { - "id": "mistral.mistral-large-3-675b-instruct", - "name": "Mistral Large 3", - "family": "mistral", + "aion-labs/aion-1.0": { + "id": "aion-labs/aion-1.0", + "name": "AionLabs: Aion-1.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2025-02-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 4, + "output": 8 } }, - "qwen.qwen3-235b-a22b-2507-v1:0": { - "id": "qwen.qwen3-235b-a22b-2507-v1:0", - "name": "Qwen3 235B A22B 2507", - "family": "qwen", + "inflection/inflection-3-pi": { + "id": "inflection/inflection-3-pi", + "name": "Inflection: Inflection 3 Pi", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-18", - "last_updated": "2025-09-18", + "release_date": "2024-10-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 8000, + "output": 1024 }, "cost": { - "input": 0.22, - "output": 0.88 + "input": 2.5, + "output": 10 } }, - "writer.palmyra-x4-v1:0": { - "id": "writer.palmyra-x4-v1:0", - "name": "Palmyra X4", - "family": "palmyra", + "inflection/inflection-3-productivity": { + "id": "inflection/inflection-3-productivity", + "name": "Inflection: Inflection 3 Productivity", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "release_date": "2024-10-11", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 122880, - "output": 8192 + "context": 8000, + "output": 1024 }, "cost": { "input": 2.5, "output": 10 } }, - "anthropic.claude-opus-4-1-20250805-v1:0": { - "id": "anthropic.claude-opus-4-1-20250805-v1:0", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "sao10k/l3.1-euryale-70b": { + "id": "sao10k/l3.1-euryale-70b", + "name": "Sao10K: Llama 3.1 Euryale 70B v2.2", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2024-08-28", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.85, + "output": 0.85 } }, - "us.deepseek.r1-v1:0": { - "id": "us.deepseek.r1-v1:0", - "name": "DeepSeek-R1 (US)", - "family": "deepseek-thinking", + "sao10k/l3.3-euryale-70b": { + "id": "sao10k/l3.3-euryale-70b", + "name": "Sao10K: Llama 3.3 Euryale 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-29", + "release_date": "2024-12-18", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.65, + "output": 0.75 } }, - "eu.anthropic.claude-opus-4-6-v1": { - "id": "eu.anthropic.claude-opus-4-6-v1", - "name": "Claude Opus 4.6 (EU)", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "sao10k/l3-lunaris-8b": { + "id": "sao10k/l3-lunaris-8b", + "name": "Sao10K: Llama 3 8B Lunaris", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2024-08-13", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.04, + "output": 0.05 } }, - "us.meta.llama4-maverick-17b-instruct-v1:0": { - "id": "us.meta.llama4-maverick-17b-instruct-v1:0", - "name": "Llama 4 Maverick 17B Instruct (US)", - "family": "llama", - "attachment": true, + "sao10k/l3-euryale-70b": { + "id": "sao10k/l3-euryale-70b", + "name": "Sao10k: Llama 3 Euryale 70B v2.1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2024-06-18", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 16384 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.24, - "output": 0.97 + "input": 1.48, + "output": 1.48 } }, - "au.anthropic.claude-haiku-4-5-20251001-v1:0": { - "id": "au.anthropic.claude-haiku-4-5-20251001-v1:0", - "name": "Claude Haiku 4.5 (AU)", - "family": "claude-haiku", - "attachment": true, - "reasoning": true, - "tool_call": true, + "sao10k/l3.1-70b-hanami-x1": { + "id": "sao10k/l3.1-70b-hanami-x1", + "name": "Sao10K: Llama 3.1 70B Hanami x1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2025-01-08", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 16000, + "output": 16000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 3, + "output": 3 } }, - "jp.anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "jp.anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5 (JP)", - "family": "claude-sonnet", - "attachment": true, + "upstage/solar-pro-3": { + "id": "upstage/solar-pro-3", + "name": "Upstage: Solar Pro 3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-01-27", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.15, + "output": 0.6 } }, - "anthropic.claude-sonnet-4-6": { - "id": "anthropic.claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "allenai/olmo-3-32b-think": { + "id": "allenai/olmo-3-32b-think", + "name": "AllenAI: Olmo 3 32B Think", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2025-11-22", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.15, + "output": 0.5 } }, - "jp.anthropic.claude-sonnet-4-6": { - "id": "jp.anthropic.claude-sonnet-4-6", - "name": "Claude Sonnet 4.6 (JP)", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "essentialai/rnj-1-instruct": { + "id": "essentialai/rnj-1-instruct", + "name": "EssentialAI: Rnj 1 Instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2025-12-05", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 32768, + "output": 6554 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.15, + "output": 0.15 } }, - "global.anthropic.claude-sonnet-4-6": { - "id": "global.anthropic.claude-sonnet-4-6", - "name": "Claude Sonnet 4.6 (Global)", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-r1-0528": { + "id": "deepseek/deepseek-r1-0528", + "name": "DeepSeek: R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2025-05-28", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 163840, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.45, + "output": 2.15, + "cache_read": 0.2 } }, - "us.anthropic.claude-sonnet-4-6": { - "id": "us.anthropic.claude-sonnet-4-6", - "name": "Claude Sonnet 4.6 (US)", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek: DeepSeek V4 Flash", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2026-04-24", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "global.anthropic.claude-opus-4-6-v1": { - "id": "global.anthropic.claude-opus-4-6-v1", - "name": "Claude Opus 4.6 (Global)", - "family": "claude-opus", - "attachment": true, + "deepseek/deepseek-v3.1-terminus": { + "id": "deepseek/deepseek-v3.1-terminus", + "name": "DeepSeek: DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 163840, + "output": 32768 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.21, + "output": 0.79, + "cache_read": 0.13 } }, - "us.anthropic.claude-opus-4-6-v1": { - "id": "us.anthropic.claude-opus-4-6-v1", - "name": "Claude Opus 4.6 (US)", - "family": "claude-opus", - "attachment": true, + "deepseek/deepseek-r1-distill-llama-70b": { + "id": "deepseek/deepseek-r1-distill-llama-70b", + "name": "DeepSeek: R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2025-01-23", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.7, + "output": 0.8, + "cache_read": 0.015 } }, - "us.anthropic.claude-opus-4-1-20250805-v1:0": { - "id": "us.anthropic.claude-opus-4-1-20250805-v1:0", - "name": "Claude Opus 4.1 (US)", - "family": "claude-opus", - "attachment": true, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek: DeepSeek V4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-04-24", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "au.anthropic.claude-sonnet-4-5-20250929-v1:0": { - "id": "au.anthropic.claude-sonnet-4-5-20250929-v1:0", - "name": "Claude Sonnet 4.5 (AU)", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-r1": { + "id": "deepseek/deepseek-r1", + "name": "DeepSeek: R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 64000, + "output": 16000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.7, + "output": 2.5 } }, - "eu.anthropic.claude-sonnet-4-6": { - "id": "eu.anthropic.claude-sonnet-4-6", - "name": "Claude Sonnet 4.6 (EU)", - "family": "claude-sonnet", - "attachment": true, + "deepseek/deepseek-v3.2-speciale": { + "id": "deepseek/deepseek-v3.2-speciale", + "name": "DeepSeek: DeepSeek V3.2 Speciale", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2025-12-01", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.4, + "output": 1.2, + "cache_read": 0.135 } }, - "us.meta.llama4-scout-17b-instruct-v1:0": { - "id": "us.meta.llama4-scout-17b-instruct-v1:0", - "name": "Llama 4 Scout 17B Instruct (US)", - "family": "llama", - "attachment": true, - "reasoning": false, + "deepseek/deepseek-v3.2-exp": { + "id": "deepseek/deepseek-v3.2-exp", + "name": "DeepSeek: DeepSeek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-01-01", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 3500000, - "output": 16384 + "context": 163840, + "output": 65536 }, "cost": { - "input": 0.17, - "output": 0.66 + "input": 0.27, + "output": 0.41 } }, - "anthropic.claude-opus-4-6-v1": { - "id": "anthropic.claude-opus-4-6-v1", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "deepseek/deepseek-chat-v3-0324": { + "id": "deepseek/deepseek-chat-v3-0324", + "name": "DeepSeek: DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2025-03-24", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 163840, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.2, + "output": 0.77, + "cache_read": 0.095 } }, - "jp.anthropic.claude-opus-4-7": { - "id": "jp.anthropic.claude-opus-4-7", - "name": "Claude Opus 4.7 (JP)", - "family": "claude-opus", - "attachment": true, + "deepseek/deepseek-r1-distill-qwen-32b": { + "id": "deepseek/deepseek-r1-distill-qwen-32b", + "name": "DeepSeek: R1 Distill Qwen 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "reasoning_options": [], + "tool_call": false, + "temperature": true, + "release_date": "2025-01-01", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.29, + "output": 0.29 } - } - } - }, - "the-grid-ai": { - "id": "the-grid-ai", - "env": ["THEGRIDAI_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.thegrid.ai/v1", - "name": "The Grid AI", - "doc": "https://thegrid.ai/docs", - "models": { - "text-prime": { - "id": "text-prime", - "name": "Text Prime", + }, + "deepseek/deepseek-chat": { + "id": "deepseek/deepseek-chat", + "name": "DeepSeek: DeepSeek V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "release_date": "2024-12-01", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 30000 + "context": 163840, + "output": 163840 }, - "status": "beta" + "cost": { + "input": 0.32, + "output": 0.89, + "cache_read": 0.15 + } }, - "text-standard": { - "id": "text-standard", - "name": "Text Standard", + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "DeepSeek: DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "release_date": "2025-12-01", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16000 + "context": 163840, + "output": 65536 }, - "status": "beta" + "cost": { + "input": 0.26, + "output": 0.38, + "cache_read": 0.125 + } }, - "text-max": { - "id": "text-max", - "name": "Text Max", + "deepseek/deepseek-chat-v3.1": { + "id": "deepseek/deepseek-chat-v3.1", + "name": "DeepSeek: DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-03-24", - "last_updated": "2026-03-24", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 32768, + "output": 7168 }, - "status": "beta" - } - } - }, - "baseten": { - "id": "baseten", - "env": ["BASETEN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://inference.baseten.co/v1", - "name": "Baseten", - "doc": "https://docs.baseten.co/development/model-apis/overview", - "models": { - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", - "name": "GLM-4.7", - "family": "glm", + "cost": { + "input": 0.15, + "output": 0.75 + } + }, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax: MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2026-02-12", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 196608, + "output": 196608 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.25, + "output": 1.2, + "cache_read": 0.029 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", - "family": "glm", + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax: MiniMax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 196608, + "output": 39322 }, "cost": { - "input": 0.95, - "output": 3.15 + "input": 0.27, + "output": 0.95, + "cache_read": 0.03 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "GLM 4.6", - "family": "glm", - "attachment": false, + "minimax/minimax-01": { + "id": "minimax/minimax-01", + "name": "MiniMax: MiniMax-01", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2025-09-16", - "last_updated": "2025-09-16", + "release_date": "2025-01-15", + "last_updated": "2025-01-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 1000192, + "output": 1000192 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.2, + "output": 1.1 } }, - "nvidia/Nemotron-120B-A12B": { - "id": "nvidia/Nemotron-120B-A12B", - "name": "Nemotron 3 Super", - "family": "nemotron", - "attachment": false, + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax: MiniMax M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2026-02", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "release_date": "2026-06-01", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32678 + "context": 1048576, + "output": 512000 }, "cost": { "input": 0.3, - "output": 0.75 + "output": 1.2, + "cache_read": 0.06 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax: MiniMax M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2025-08-25", - "last_updated": "2025-08-25", + "release_date": "2025-10-23", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 164000, - "output": 131000 + "context": 196608, + "output": 196608 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.255, + "output": 1, + "cache_read": 0.03 } }, - "deepseek-ai/DeepSeek-V3-0324": { - "id": "deepseek-ai/DeepSeek-V3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", + "minimax/minimax-m2-her": { + "id": "minimax/minimax-m2-her", + "name": "MiniMax: MiniMax M2-her", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "release_date": "2026-01-23", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 164000, - "output": 131000 + "context": 65536, + "output": 2048 }, "cost": { - "input": 0.77, - "output": 0.77 + "input": 0.3, + "output": 1.2 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax: MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax-m2.7", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-10", - "release_date": "2025-12-01", - "last_updated": "2026-03-06", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163800, - "output": 131100 + "context": 204800, + "output": 131072 }, - "status": "deprecated", "cost": { "input": 0.3, - "output": 0.45 + "output": 1.2, + "cache_read": 0.06 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "minimax/minimax-m1": { + "id": "minimax/minimax-m1", + "name": "MiniMax: MiniMax M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 1000000, + "output": 40000 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.4, + "output": 2.2 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "stealth/claude-opus-4.7": { + "id": "stealth/claude-opus-4.7", + "name": "Stealth: Claude Opus 4.7 (20% off)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "temperature": false, + "release_date": "2026-04-16", + "last_updated": "2026-05-27", + "modalities": { + "input": ["image", "pdf", "text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 }, + "cost": { + "input": 4, + "output": 20, + "cache_read": 0.4, + "cache_write": 5 + } + }, + "stealth/claude-sonnet-4.6": { + "id": "stealth/claude-sonnet-4.6", + "name": "Stealth: Claude Sonnet 4.6 (20% off)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2026-03-06", + "release_date": "2026-02-17", + "last_updated": "2026-05-27", "modalities": { - "input": ["text"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 64000 }, - "status": "deprecated", "cost": { - "input": 0.6, - "output": 2.5 + "input": 2.4, + "output": 12, + "cache_read": 0.24, + "cache_write": 3 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "stealth/claude-opus-4.6": { + "id": "stealth/claude-opus-4.6", + "name": "Stealth: Claude Opus 4.6 (20% off)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-02-05", + "last_updated": "2026-05-27", "modalities": { - "input": ["text", "image", "video"], + "input": ["image", "pdf", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 4, + "output": 20, + "cache_read": 0.4, + "cache_write": 5 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi K2 Instruct 0905", - "family": "kimi", + "kwaipilot/kat-coder-pro-v2": { + "id": "kwaipilot/kat-coder-pro-v2", + "name": "Kwaipilot: KAT-Coder-Pro V2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-09-05", - "last_updated": "2026-03-06", + "release_date": "2026-03-27", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 256000, + "output": 80000 }, - "status": "deprecated", "cost": { - "input": 0.6, - "output": 2.5 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "nousresearch/hermes-2-pro-llama-3-8b": { + "id": "nousresearch/hermes-2-pro-llama-3-8b", + "name": "NousResearch: Hermes 2 Pro - Llama-3 8B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-01-30", - "last_updated": "2026-02-12", + "release_date": "2024-05-27", + "last_updated": "2024-06-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 8192, "output": 8192 }, "cost": { - "input": 0.6, - "output": 3 + "input": 0.14, + "output": 0.14 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "nousresearch/hermes-4-405b": { + "id": "nousresearch/hermes-4-405b", + "name": "Nous: Hermes 4 405B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-08-25", + "last_updated": "2025-08-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204000, - "output": 204000 + "context": 131072, + "output": 26215 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1, + "output": 3 } }, - "deepseek-ai/DeepSeek-V4-Pro": { - "id": "deepseek-ai/DeepSeek-V4-Pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "nousresearch/hermes-3-llama-3.1-70b": { + "id": "nousresearch/hermes-3-llama-3.1-70b", + "name": "Nous: Hermes 3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2024-08-18", + "last_updated": "2026-03-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.15 + "input": 0.3, + "output": 0.3 } - } - } - }, - "frogbot": { - "id": "frogbot", - "env": ["FROGBOT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://app.frogbot.ai/api/v1", - "name": "FrogBot", - "doc": "https://docs.frogbot.ai", - "models": { - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "Grok 4.1 Fast (Reasoning)", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": true, + }, + "nousresearch/hermes-3-llama-3.1-405b": { + "id": "nousresearch/hermes-3-llama-3.1-405b", + "name": "Nous: Hermes 3 405B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "release_date": "2024-08-16", + "last_updated": "2024-08-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 1, + "output": 1 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, + "nousresearch/hermes-4-70b": { + "id": "nousresearch/hermes-4-70b", + "name": "Nous: Hermes 4 70B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2025-08-25", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.13, + "output": 0.4, + "cache_read": 0.055 } - }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi-K2.5", - "attachment": false, + } + } + }, + "lucidquery": { + "id": "lucidquery", + "env": ["LUCIDQUERY_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.lucidquery.com/v1", + "name": "LucidQuery", + "doc": "https://lucidquery.com/docs", + "models": { + "lucidnova-rf1-100b": { + "id": "lucidnova-rf1-100b", + "name": "LucidNova RF1 100B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "nova", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "1970-01-01", - "last_updated": "1970-01-01", + "temperature": false, + "knowledge": "2025-09-16", + "release_date": "2024-12-28", + "last_updated": "2025-09-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 128000 + "context": 120000, + "output": 8000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 2, + "output": 5 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "lucidquery-nexus-coder": { + "id": "lucidquery-nexus-coder", + "name": "LucidQuery Nexus Coder", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "lucid", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "temperature": false, + "knowledge": "2025-08-01", + "release_date": "2025-09-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 250000, + "output": 60000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 2, + "output": 5 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "lucidquery-agi-01-swift": { + "id": "lucidquery-agi-01-swift", + "name": "AGI-01 Swift", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "agi", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "knowledge": "2026-06-05", + "release_date": "2026-06-16", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 300000, + "output": 120000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 + "input": 2.5, + "output": 15 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "lucidquery-agi-01-frontier": { + "id": "lucidquery-agi-01-frontier", + "name": "AGI-01 Frontier", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "agi", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "knowledge": "2026-06-05", + "release_date": "2026-06-16", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 300000, + "output": 120000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 4.5, + "output": 22 } - }, - "zai-glm-5-1": { - "id": "zai-glm-5-1", - "name": "Z.AI GLM-5.1", - "family": "glm", - "attachment": true, + } + } + }, + "meganova": { + "id": "meganova", + "env": ["MEGANOVA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.meganova.ai/v1", + "name": "Meganova", + "doc": "https://docs.meganova.ai", + "models": { + "meta-llama/Llama-3.3-70B-Instruct": { + "id": "meta-llama/Llama-3.3-70B-Instruct", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-01-20", - "last_updated": "2025-02-22", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, - "output": 8192 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.1, + "output": 0.3 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, + "moonshotai/Kimi-K2-Thinking": { + "id": "moonshotai/Kimi-K2-Thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 0.6, + "output": 2.6 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, - "reasoning": false, + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "knowledge": "2026-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.45, + "output": 2.8 } }, - "gpt-5-4-nano": { - "id": "gpt-5-4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", + "Qwen/Qwen2.5-VL-32B-Instruct": { + "id": "Qwen/Qwen2.5-VL-32B-Instruct", + "name": "Qwen2.5 VL 32B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 16384, + "output": 16384 }, "cost": { "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "output": 0.6 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, + "Qwen/Qwen3.5-Plus": { + "id": "Qwen/Qwen3.5-Plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-07-17", - "last_updated": "2025-07-17", + "knowledge": "2025-04", + "release_date": "2026-02", + "last_updated": "2026-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1000000, "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075 + "input": 0.4, + "output": 2.4, + "reasoning": 2.4 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok 4.1 Fast (Reasoning)", - "family": "grok", + "Qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.09, + "output": 0.6 } }, - "gpt-5-5": { - "id": "gpt-5-5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "XiaomiMiMo/MiMo-V2-Flash": { + "id": "XiaomiMiMo/MiMo-V2-Flash", + "name": "MiMo V2 Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "knowledge": "2024-12-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 272000, - "output": 128000 + "context": 262144, + "output": 32000 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.1, + "output": 0.3 } }, - "grok-4-3": { - "id": "grok-4-3", - "name": "Grok 4.3", - "family": "grok", + "mistralai/Mistral-Small-3.2-24B-Instruct-2506": { + "id": "mistralai/Mistral-Small-3.2-24B-Instruct-2506", + "name": "Mistral Small 3.2 24B Instruct", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2026-04-30", - "last_updated": "2026-04-30", + "knowledge": "2024-10", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 32768, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2 + "input": 0, + "output": 0 } }, - "gpt-5-4-mini": { - "id": "gpt-5-4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "mistralai/Mistral-Nemo-Instruct-2407": { + "id": "mistralai/Mistral-Nemo-Instruct-2407", + "name": "Mistral Nemo Instruct 2407", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0.02, + "output": 0.04 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "zai-org/GLM-4.6": { + "id": "zai-org/GLM-4.6", + "name": "GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 128000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } - }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek v4 Pro", - "family": "deepseek", - "attachment": true, - "reasoning": false, - "tool_call": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 202752, + "output": 131072 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.14 + "input": 0.45, + "output": 1.9 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "1970-01-01", - "last_updated": "1970-01-01", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.07, - "output": 0.2 + "input": 0.8, + "output": 2.56 } }, - "qwen-3-6-plus": { - "id": "qwen-3-6-plus", - "name": "Qwen 3.6 Plus", - "family": "qwen", - "attachment": true, + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-03", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.1 + "input": 0.2, + "output": 0.8 } }, - "minimax-m2-7": { - "id": "minimax-m2-7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": true, + "deepseek-ai/DeepSeek-V3-0324": { + "id": "deepseek-ai/DeepSeek-V3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 192000, - "output": 8192 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0.25, + "output": 0.88 } }, - "minimax-m2-5": { - "id": "minimax-m2-5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": true, - "reasoning": false, - "tool_call": true, + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-01-15", - "last_updated": "2025-02-22", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 192000, - "output": 8192 + "context": 163840, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03 + "input": 0.5, + "output": 2.15 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "release_date": "2025-08-25", + "last_updated": "2025-08-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 164000, + "output": 164000 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.27, + "output": 1 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "deepseek-ai/DeepSeek-V3.2-Exp": { + "id": "deepseek-ai/DeepSeek-V3.2-Exp", + "name": "DeepSeek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "1970-01-01", - "last_updated": "1970-01-01", + "release_date": "2025-10-10", + "last_updated": "2025-10-10", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 164000, + "output": 164000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.27, + "output": 0.4 } }, - "gemini-3-1-pro-preview": { - "id": "gemini-3-1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "deepseek-ai/DeepSeek-V3.2": { + "id": "deepseek-ai/DeepSeek-V3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-18", - "last_updated": "2026-02-18", + "release_date": "2025-12-03", + "last_updated": "2025-12-03", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 164000, + "output": 164000 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2 + "input": 0.26, + "output": 0.38 } }, - "kimi-k2-6": { - "id": "kimi-k2-6", - "name": "Kimi-K2.6", + "MiniMaxAI/MiniMax-M2.1": { + "id": "MiniMaxAI/MiniMax-M2.1", + "name": "MiniMax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "1970-01-01", - "last_updated": "1970-01-01", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 196608, + "output": 131072 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.28, + "output": 1.2 } }, - "gpt-5-3-codex": { - "id": "gpt-5-3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt", - "attachment": true, + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.3, + "output": 1.2 } } } }, - "zhipuai-coding-plan": { - "id": "zhipuai-coding-plan", - "env": ["ZHIPU_API_KEY"], + "inferx": { + "id": "inferx", + "env": ["INFERX_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://open.bigmodel.cn/api/coding/paas/v4", - "name": "Zhipu AI Coding Plan", - "doc": "https://docs.bigmodel.cn/cn/coding-plan/overview", + "api": "https://model.inferx.net/v1", + "name": "InferX", + "doc": "https://model.inferx.net/endpoints", "models": { - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM-5V-Turbo", - "family": "glm", + "google/gemma-4-31b-it-fp8": { + "id": "google/gemma-4-31b-it-fp8", + "name": "Gemma 4 31B IT FP8", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 262144, + "output": 32768 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "qwen/qwen3-coder-next-fp8-1m": { + "id": "qwen/qwen3-coder-next-fp8-1m", + "name": "Qwen3 Coder Next FP8 1M", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2025-04", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 1024000, + "output": 65536 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "glm-5-turbo": { - "id": "glm-5-turbo", - "name": "GLM-5-Turbo", - "family": "glm", - "attachment": false, + "qwen/qwen3.5-122b-a10b-nvfp4": { + "id": "qwen/qwen3.5-122b-a10b-nvfp4", + "name": "Qwen3.5 122B A10B NVFP4", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 256144, + "output": 65536 }, + "cost": { + "input": 0, + "output": 0 + } + }, + "qwen/qwen3.6-35b-a3b-fp8": { + "id": "qwen/qwen3.6-35b-a3b-fp8", + "name": "Qwen3.6 35B A3B FP8", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 262000, + "output": 65536 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM-4.5-Air", - "family": "glm-air", + "qwen/qwen3-coder-next-fp8": { + "id": "qwen/qwen3-coder-next-fp8", + "name": "Qwen3 Coder Next FP8", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 256144, + "output": 65536 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "qwen/qwen3.6-27b-fp8": { + "id": "qwen/qwen3.6-27b-fp8", + "name": "Qwen3.6 27B FP8", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } } } }, - "alibaba-coding-plan": { - "id": "alibaba-coding-plan", - "env": ["ALIBABA_CODING_PLAN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://coding-intl.dashscope.aliyuncs.com/v1", - "name": "Alibaba Coding Plan", - "doc": "https://www.alibabacloud.com/help/en/model-studio/coding-plan", + "perplexity": { + "id": "perplexity", + "env": ["PERPLEXITY_API_KEY"], + "npm": "@ai-sdk/perplexity", + "name": "Perplexity", + "doc": "https://docs.perplexity.ai", "models": { - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3 Coder Plus", - "family": "qwen", + "sonar-reasoning-pro": { + "id": "sonar-reasoning-pro", + "name": "Sonar Reasoning Pro", + "description": "Web-grounded Sonar for multi-step research questions that need cited reasoning", + "family": "sonar-reasoning", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": true, + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 4096 + }, + "cost": { + "input": 2, + "output": 8 + } + }, + "sonar": { + "id": "sonar", + "name": "Sonar", + "description": "Fast web-grounded Sonar for current answers, citations, and lightweight retrieval", + "family": "sonar", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1, + "output": 1 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "sonar-pro": { + "id": "sonar-pro", + "name": "Sonar Pro", + "description": "Deeper Sonar search model with broader retrieval and stronger synthesis", + "family": "sonar-pro", "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2025-09-01", + "release_date": "2024-01-01", + "last_updated": "2025-09-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 200000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 3, + "output": 15 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "sonar-deep-research": { + "id": "sonar-deep-research", + "name": "Perplexity Sonar Deep Research", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-02-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 2, + "output": 8, + "reasoning": 3 } - }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, + } + } + }, + "amazon-bedrock": { + "id": "amazon-bedrock", + "env": ["AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY", "AWS_REGION", "AWS_BEARER_TOKEN_BEDROCK"], + "npm": "@ai-sdk/amazon-bedrock", + "name": "Amazon Bedrock", + "doc": "https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html", + "models": { + "global.anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "global.anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5 (Global)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, + "global.anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "global.anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5 (Global)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "input": 196601, - "output": 24576 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen", - "attachment": false, + "jp.anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "jp.anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5 (JP)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "qwen3-max-2026-01-23": { - "id": "qwen3-max-2026-01-23", - "name": "Qwen3 Max", - "family": "qwen", - "attachment": false, + "us.meta.llama4-scout-17b-instruct-v1:0": { + "id": "us.meta.llama4-scout-17b-instruct-v1:0", + "name": "Llama 4 Scout 17B Instruct (US)", + "description": "Open Llama with long-context vision for efficient multimodal agents", + "family": "llama", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-23", - "last_updated": "2026-01-23", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 3500000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.17, + "output": 0.66 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "Qwen3 Coder Next", - "family": "qwen", + "minimax.minimax-m2": { + "id": "minimax.minimax-m2", + "name": "MiniMax M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-02-03", - "last_updated": "2026-02-03", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 204608, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.3, + "output": 1.2 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", - "family": "qwen", - "attachment": false, + "anthropic.claude-opus-4-7": { + "id": "anthropic.claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 65536 + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } - } - } - }, - "venice": { - "id": "venice", - "env": ["VENICE_API_KEY"], - "npm": "venice-ai-sdk-provider", - "name": "Venice AI", - "doc": "https://docs.venice.ai", - "models": { - "openai-gpt-4o-mini-2024-07-18": { - "id": "openai-gpt-4o-mini-2024-07-18", - "name": "GPT-4o Mini", - "family": "gpt-mini", + }, + "eu.anthropic.claude-sonnet-4-6": { + "id": "eu.anthropic.claude-sonnet-4-6", + "name": "Claude Sonnet 4.6 (EU)", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-28", - "last_updated": "2026-03-06", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.1875, - "output": 0.75, - "cache_read": 0.09375 + "input": 3.3, + "output": 16.5, + "cache_read": 0.33, + "cache_write": 4.125 } }, - "qwen3-next-80b": { - "id": "qwen3-next-80b", - "name": "Qwen 3 Next 80b", - "family": "qwen", - "attachment": false, + "mistral.voxtral-small-24b-2507": { + "id": "mistral.voxtral-small-24b-2507", + "name": "Voxtral Small 24B 2507", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral", + "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-04-29", - "last_updated": "2026-03-12", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { - "input": ["text"], + "input": ["text", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 32000, + "output": 8192 }, "cost": { - "input": 0.35, - "output": 1.9 + "input": 0.15, + "output": 0.35 } }, - "grok-4-20-multi-agent": { - "id": "grok-4-20-multi-agent", - "name": "Grok 4.20 Multi-Agent", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": false, + "mistral.ministral-3-3b-instruct": { + "id": "mistral.ministral-3-3b-instruct", + "name": "Ministral 3 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": false, + "reasoning": false, + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-12", - "last_updated": "2026-05-07", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 128000 + "context": 256000, + "output": 8192 }, "cost": { - "input": 1.42, - "output": 2.83, - "cache_read": 0.23, - "tiers": [ - { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45 - } + "input": 0.1, + "output": 0.1 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "Qwen 3 235B A22B Instruct 2507", - "family": "qwen", + "openai.gpt-oss-20b": { + "id": "openai.gpt-oss-20b", + "name": "gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-04-29", - "last_updated": "2026-03-12", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, "output": 16384 }, + "provider": { + "npm": "@ai-sdk/amazon-bedrock/mantle", + "api": "https://bedrock-mantle.${AWS_REGION}.api.aws/v1", + "shape": "responses" + }, "cost": { - "input": 0.15, - "output": 0.75 + "input": 0.07, + "output": 0.3 } }, - "z-ai-glm-5v-turbo": { - "id": "z-ai-glm-5v-turbo", - "name": "GLM 5V Turbo", - "family": "glmv", + "anthropic.claude-opus-4-6-v1": { + "id": "anthropic.claude-opus-4-6-v1", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-12", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1.5, - "output": 5, - "cache_read": 0.3 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "gemma-4-uncensored": { - "id": "gemma-4-uncensored", - "name": "Gemma 4 Uncensored", - "family": "gemma", - "attachment": true, + "openai.gpt-oss-safeguard-20b": { + "id": "openai.gpt-oss-safeguard-20b", + "name": "GPT OSS Safeguard 20B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", + "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-13", - "last_updated": "2026-04-19", + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1625, - "output": 0.5 + "input": 0.07, + "output": 0.2 } }, - "grok-41-fast": { - "id": "grok-41-fast", - "name": "Grok 4.1 Fast", - "family": "grok", + "anthropic.claude-opus-4-5-20251101-v1:0": { + "id": "anthropic.claude-opus-4-5-20251101-v1:0", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-12-01", - "last_updated": "2026-04-09", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 30000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.23, - "output": 0.57, - "cache_read": 0.06 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "global.anthropic.claude-fable-5": { + "id": "global.anthropic.claude-fable-5", + "name": "Claude Fable 5 (Global)", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-16", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 64000 + "output": 128000 }, "cost": { - "input": 3.6, - "output": 18, - "cache_read": 0.36, - "cache_write": 4.5 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "nvidia-nemotron-cascade-2-30b-a3b": { - "id": "nvidia-nemotron-cascade-2-30b-a3b", - "name": "Nemotron Cascade 2 30B A3B", - "family": "nemotron", + "openai.gpt-oss-120b-1:0": { + "id": "openai.gpt-oss-120b-1:0", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-24", - "last_updated": "2026-04-09", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.8 + "input": 0.15, + "output": 0.6 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-19", - "last_updated": "2026-03-12", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.7, - "output": 3.75, - "cache_read": 0.07 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "grok-4-20": { - "id": "grok-4-20", - "name": "Grok 4.20", - "family": "grok", + "amazon.nova-pro-v1:0": { + "id": "amazon.nova-pro-v1:0", + "name": "Nova Pro", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova-pro", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-03-12", - "last_updated": "2026-05-07", + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 128000 + "context": 300000, + "output": 8192 }, "cost": { - "input": 1.42, - "output": 2.83, - "cache_read": 0.23, - "tiers": [ - { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45 - } + "input": 0.8, + "output": 3.2, + "cache_read": 0.2 } }, - "google-gemma-4-26b-a4b-it": { - "id": "google-gemma-4-26b-a4b-it", - "name": "Google Gemma 4 26B A4B Instruct", - "family": "gemma", - "attachment": true, + "qwen.qwen3-coder-next": { + "id": "qwen.qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-12", + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.1625, - "output": 0.5 + "input": 0.22, + "output": 1.8 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", + "us.anthropic.claude-opus-4-7": { + "id": "us.anthropic.claude-opus-4-7", + "name": "Claude Opus 4.7 (US)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, + "knowledge": "2026-01-31", "release_date": "2026-04-16", "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, @@ -71288,388 +119105,444 @@ "output": 128000 }, "cost": { - "input": 6, - "output": 30, - "cache_read": 0.6, - "cache_write": 7.5 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen3-coder-480b-a35b-instruct-turbo": { - "id": "qwen3-coder-480b-a35b-instruct-turbo", - "name": "Qwen 3 Coder 480B Turbo", - "family": "qwen", + "nvidia.nemotron-nano-9b-v2": { + "id": "nvidia.nemotron-nano-9b-v2", + "name": "NVIDIA Nemotron Nano 9B v2", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-02-26", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 256000, - "output": 65536 - }, - "cost": { - "input": 0.35, - "output": 1.5, - "cache_read": 0.04 - } - }, - "qwen3-5-397b-a17b": { - "id": "qwen3-5-397b-a17b", - "name": "Qwen 3.5 397B", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 32768 + "output": 4096 }, "cost": { - "input": 0.75, - "output": 4.5 + "input": 0.06, + "output": 0.23 } }, - "zai-org-glm-4.7": { - "id": "zai-org-glm-4.7", - "name": "GLM 4.7", - "family": "glm", + "qwen.qwen3-32b-v1:0": { + "id": "qwen.qwen3-32b-v1:0", + "name": "Qwen3 32B (dense)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-24", - "last_updated": "2026-03-12", + "knowledge": "2024-04", + "release_date": "2025-09-18", + "last_updated": "2025-09-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, + "context": 16384, "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.65, - "cache_read": 0.11 + "input": 0.15, + "output": 0.6 } }, - "openai-gpt-54": { - "id": "openai-gpt-54", - "name": "GPT-5.4", - "family": "gpt", + "jp.anthropic.claude-sonnet-4-6": { + "id": "jp.anthropic.claude-sonnet-4-6", + "name": "Claude Sonnet 4.6 (JP)", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-05", - "last_updated": "2026-03-09", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 131072 + "output": 64000 }, "cost": { - "input": 3.13, - "output": 18.8, - "cache_read": 0.313 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "zai-org-glm-4.7-flash": { - "id": "zai-org-glm-4.7-flash", - "name": "GLM 4.7 Flash", - "family": "glm-flash", + "deepseek.r1-v1:0": { + "id": "deepseek.r1-v1:0", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-01-29", - "last_updated": "2026-03-12", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "output": 32768 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 1.35, + "output": 5.4 } }, - "nvidia-nemotron-3-nano-30b-a3b": { - "id": "nvidia-nemotron-3-nano-30b-a3b", - "name": "NVIDIA Nemotron 3 Nano 30B", - "family": "nemotron", + "mistral.mistral-large-3-675b-instruct": { + "id": "mistral.mistral-large-3-675b-instruct", + "name": "Mistral Large 3", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-03-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 256000, + "output": 8192 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.5, + "output": 1.5 } }, - "qwen3-vl-235b-a22b": { - "id": "qwen3-vl-235b-a22b", - "name": "Qwen3 VL 235B", - "family": "qwen", + "google.gemma-3-27b-it": { + "id": "google.gemma-3-27b-it", + "name": "Google Gemma 3 27B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01-16", - "last_updated": "2026-03-12", + "knowledge": "2025-07", + "release_date": "2025-07-27", + "last_updated": "2025-07-27", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 202752, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 1.5 + "input": 0.12, + "output": 0.2 } }, - "openai-gpt-53-codex": { - "id": "openai-gpt-53-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", + "anthropic.claude-sonnet-4-6": { + "id": "anthropic.claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-03-12", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 2.19, - "output": 17.5, - "cache_read": 0.219 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "venice-uncensored-1-2": { - "id": "venice-uncensored-1-2", - "name": "Venice Uncensored 1.2", - "family": "venice", - "attachment": true, - "reasoning": false, + "amazon.nova-2-lite-v1:0": { + "id": "amazon.nova-2-lite-v1:0", + "name": "Nova 2 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "nova", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-19", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "output": 4096 }, "cost": { - "input": 0.2, - "output": 0.9 + "input": 0.33, + "output": 2.75 } }, - "openai-gpt-52": { - "id": "openai-gpt-52", - "name": "GPT-5.2", - "family": "gpt", + "openai.gpt-oss-safeguard-120b": { + "id": "openai.gpt-oss-safeguard-120b", + "name": "GPT OSS Safeguard 120B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2025-12-13", - "last_updated": "2026-03-12", + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2.19, - "output": 17.5, - "cache_read": 0.219 + "input": 0.15, + "output": 0.6 } }, - "mistral-small-3-2-24b-instruct": { - "id": "mistral-small-3-2-24b-instruct", - "name": "Mistral Small 3.2 24B Instruct", - "family": "mistral-small", + "mistral.ministral-3-8b-instruct": { + "id": "mistral.ministral-3-8b-instruct", + "name": "Ministral 3 8B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-01-15", - "last_updated": "2026-03-16", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.09375, - "output": 0.25 + "input": 0.15, + "output": 0.15 } }, - "minimax-m27": { - "id": "minimax-m27", - "name": "MiniMax M2.7", - "family": "minimax", - "attachment": false, + "eu.anthropic.claude-opus-4-6-v1": { + "id": "eu.anthropic.claude-opus-4-6-v1", + "name": "Claude Opus 4.6 (EU)", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-04-12", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 198000, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.375, - "output": 1.5, - "cache_read": 0.075 + "input": 5.5, + "output": 27.5, + "cache_read": 0.55, + "cache_write": 6.875 } }, - "qwen3-235b-a22b-thinking-2507": { - "id": "qwen3-235b-a22b-thinking-2507", - "name": "Qwen 3 235B A22B Thinking 2507", - "family": "qwen", - "attachment": false, + "au.anthropic.claude-opus-4-6-v1": { + "id": "au.anthropic.claude-opus-4-6-v1", + "name": "AU Anthropic Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-04-29", - "last_updated": "2026-03-12", + "knowledge": "2025-05", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 3.5 + "input": 16.5, + "output": 82.5, + "cache_read": 1.65, + "cache_write": 20.625 } }, - "qwen3-5-35b-a3b": { - "id": "qwen3-5-35b-a3b", - "name": "Qwen 3.5 35B A3B", - "family": "qwen", + "jp.anthropic.claude-sonnet-5": { + "id": "jp.anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5 (JP)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-02-25", - "last_updated": "2026-04-16", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3125, - "output": 1.25, - "cache_read": 0.15625 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "mercury-2": { - "id": "mercury-2", - "name": "Mercury 2", - "family": "mercury", + "openai.gpt-oss-120b": { + "id": "openai.gpt-oss-120b", + "name": "gpt-oss-120b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-20", - "last_updated": "2026-04-09", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] @@ -71677,779 +119550,819 @@ "open_weights": false, "limit": { "context": 128000, - "output": 50000 + "output": 16384 + }, + "provider": { + "npm": "@ai-sdk/amazon-bedrock/mantle", + "api": "https://bedrock-mantle.${AWS_REGION}.api.aws/v1", + "shape": "responses" }, "cost": { - "input": 0.3125, - "output": 0.9375, - "cache_read": 0.03125 + "input": 0.15, + "output": 0.6 } }, - "google-gemma-3-27b-it": { - "id": "google-gemma-3-27b-it", - "name": "Google Gemma 3 27B Instruct", - "family": "gemma", + "global.anthropic.claude-sonnet-5": { + "id": "global.anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5 (Global)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-04", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 198000, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0.2 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "olafangensan-glm-4.7-flash-heretic": { - "id": "olafangensan-glm-4.7-flash-heretic", - "name": "GLM 4.7 Flash Heretic", - "family": "glm-flash", - "attachment": false, + "global.anthropic.claude-opus-4-8": { + "id": "global.anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8 (Global)", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-04", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 24000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.8 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "openai-gpt-55-pro": { - "id": "openai-gpt-55-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", + "us.meta.llama4-maverick-17b-instruct-v1:0": { + "id": "us.meta.llama4-maverick-17b-instruct-v1:0", + "name": "Llama 4 Maverick 17B Instruct (US)", + "description": "Open multimodal Llama for strong reasoning with efficient everyday serving", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-25", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 1000000, - "output": 128000 + "output": 16384 }, "cost": { - "input": 37.5, - "output": 225 + "input": 0.24, + "output": 0.97 } }, - "openai-gpt-52-codex": { - "id": "openai-gpt-52-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", + "openai.gpt-5.4": { + "id": "openai.gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-01-15", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/amazon-bedrock/mantle", + "api": "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1", + "shape": "responses" }, "cost": { - "input": 2.19, - "output": 17.5, - "cache_read": 0.219 + "input": 2.75, + "output": 16.5, + "cache_read": 0.275 } }, - "venice-uncensored-role-play": { - "id": "venice-uncensored-role-play", - "name": "Venice Role Play Uncensored", - "family": "venice", + "anthropic.claude-sonnet-5": { + "id": "anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-02-20", - "last_updated": "2026-03-16", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 2 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "zai-org-glm-5": { - "id": "zai-org-glm-5", - "name": "GLM 5", - "family": "glm", + "mistral.devstral-2-123b": { + "id": "mistral.devstral-2-123b", + "name": "Devstral 2 123B", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-03-12", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, - "output": 32000 + "context": 256000, + "output": 8192 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.4, + "output": 2 } }, - "zai-org-glm-4.6": { - "id": "zai-org-glm-4.6", - "name": "GLM 4.6", + "zai.glm-4.7": { + "id": "zai.glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2024-04-01", - "last_updated": "2026-04-04", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.85, - "output": 2.75, - "cache_read": 0.3 + "input": 0.6, + "output": 2.2 } }, - "grok-4-3": { - "id": "grok-4-3", - "name": "Grok 4.3", - "family": "grok", + "anthropic.claude-fable-5": { + "id": "anthropic.claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-18", - "last_updated": "2026-05-04", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 32000 - }, - "cost": { - "input": 1.42, - "output": 2.83, - "cache_read": 0.23, - "tiers": [ - { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.83, - "output": 5.67, - "cache_read": 0.45 - } - } - }, - "mistral-small-2603": { - "id": "mistral-small-2603", - "name": "Mistral Small 4", - "family": "mistral-small", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-04-09", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 256000, - "output": 65536 + "output": 128000 }, "cost": { - "input": 0.1875, - "output": 0.75 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "openai-gpt-oss-120b": { - "id": "openai-gpt-oss-120b", - "name": "OpenAI GPT OSS 120B", - "family": "gpt-oss", + "writer.palmyra-x4-v1:0": { + "id": "writer.palmyra-x4-v1:0", + "name": "Palmyra X4", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "palmyra", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-06", - "last_updated": "2026-05-06", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 122880, + "output": 8192 }, "cost": { - "input": 0.07, - "output": 0.3 + "input": 2.5, + "output": 10 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "mistral.magistral-small-2509": { + "id": "mistral.magistral-small-2509", + "name": "Magistral Small 1.2", + "description": "Mistral reasoning model for transparent analysis, math, and complex decisions", + "family": "magistral", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-12-06", - "last_updated": "2026-04-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 198000, - "output": 32768 + "context": 128000, + "output": 40000 }, "cost": { - "input": 6, - "output": 30, - "cache_read": 0.6, - "cache_write": 7.5 + "input": 0.5, + "output": 1.5 } }, - "qwen3-5-9b": { - "id": "qwen3-5-9b", - "name": "Qwen 3.5 9B", + "qwen.qwen3-coder-480b-a35b-v1:0": { + "id": "qwen.qwen3-coder-480b-a35b-v1:0", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", - "attachment": true, - "reasoning": true, + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-05", - "last_updated": "2026-04-19", + "knowledge": "2024-04", + "release_date": "2025-09-18", + "last_updated": "2025-09-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32768 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.15 + "input": 0.22, + "output": 1.8 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "amazon.nova-micro-v1:0": { + "id": "amazon.nova-micro-v1:0", + "name": "Nova Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-micro", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-29", + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.17, - "output": 0.35, - "cache_read": 0.028 + "input": 0.035, + "output": 0.14, + "cache_read": 0.00875 } }, - "openai-gpt-54-pro": { - "id": "openai-gpt-54-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "mistral.pixtral-large-2502-v1:0": { + "id": "mistral.pixtral-large-2502-v1:0", + "name": "Pixtral Large (25.02)", + "description": "Mistral vision-language model for image understanding and multimodal chat", + "family": "mistral", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-03-05", - "last_updated": "2026-03-09", + "release_date": "2025-04-08", + "last_updated": "2025-04-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 37.5, - "output": 225, - "tiers": [ - { - "input": 75, - "output": 337.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 75, - "output": 337.5 - } + "input": 2, + "output": 6 } }, - "openai-gpt-54-mini": { - "id": "openai-gpt-54-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", + "us.anthropic.claude-opus-4-6-v1": { + "id": "us.anthropic.claude-opus-4-6-v1", + "name": "Claude Opus 4.6 (US)", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-31", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 0.9375, - "output": 5.625, - "cache_read": 0.09375 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "minimax-m25": { - "id": "minimax-m25", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "jp.anthropic.claude-opus-4-7": { + "id": "jp.anthropic.claude-opus-4-7", + "name": "Claude Opus 4.7 (JP)", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-04-12", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 198000, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.34, - "output": 1.19, - "cache_read": 0.04 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "zai-org-glm-5-1": { - "id": "zai-org-glm-5-1", - "name": "GLM 5.1", - "family": "glm", - "attachment": false, + "au.anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "au.anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5 (AU)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-08", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 200000, - "output": 24000 + "output": 64000 }, "cost": { - "input": 1.75, - "output": 5.5, - "cache_read": 0.325 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openai-gpt-55": { - "id": "openai-gpt-55", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "deepseek.v3-v1:0": { + "id": "deepseek.v3-v1:0", + "name": "DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-23", - "last_updated": "2026-04-25", + "knowledge": "2024-07", + "release_date": "2025-09-18", + "last_updated": "2025-09-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 131072 + "context": 163840, + "output": 81920 }, "cost": { - "input": 6.25, - "output": 37.5, - "cache_read": 0.625, - "tiers": [ - { - "input": 12.5, - "output": 56.25, - "cache_read": 1.25, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 12.5, - "output": 56.25, - "cache_read": 1.25 - } + "input": 0.58, + "output": 1.68 } }, - "qwen3-6-27b": { - "id": "qwen3-6-27b", - "name": "Qwen 3.6 27B", - "family": "qwen", + "anthropic.claude-opus-4-1-20250805-v1:0": { + "id": "anthropic.claude-opus-4-1-20250805-v1:0", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-29", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.325, - "output": 3.25 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "jp.anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "jp.anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5 (JP)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-16", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 6, - "output": 30, - "cache_read": 0.6, - "cache_write": 7.5 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek", + "google.gemma-3-4b-it": { + "id": "google.gemma-3-4b-it", + "name": "Gemma 3 4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-29", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 128000, + "output": 4096 }, "cost": { - "input": 1.73, - "output": 3.796, - "cache_read": 0.33 + "input": 0.04, + "output": 0.08 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, + "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "eu.anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5 (EU)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-10", - "release_date": "2025-12-04", - "last_updated": "2026-03-24", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 160000, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.33, - "output": 0.48, - "cache_read": 0.16 + "input": 3.3, + "output": 16.5, + "cache_read": 0.33, + "cache_write": 4.125 } }, - "qwen-3-6-plus": { - "id": "qwen-3-6-plus", - "name": "Qwen 3.6 Plus Uncensored", + "qwen.qwen3-vl-235b-a22b": { + "id": "qwen.qwen3-vl-235b-a22b", + "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-06", - "last_updated": "2026-04-12", + "release_date": "2025-10-04", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 0.625, - "output": 3.75, - "cache_read": 0.0625, - "cache_write": 0.78, - "tiers": [ - { - "input": 2.5, - "output": 7.5, - "cache_read": 0.0625, - "cache_write": 0.78, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 7.5, - "cache_read": 0.0625, - "cache_write": 0.78 - } + "input": 0.3, + "output": 1.5 } }, - "aion-labs-aion-2-0": { - "id": "aion-labs-aion-2-0", - "name": "Aion 2.0", - "family": "o", + "writer.palmyra-x5-v1:0": { + "id": "writer.palmyra-x5-v1:0", + "name": "Palmyra X5", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "palmyra", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2026-03-24", - "last_updated": "2026-04-12", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1040000, + "output": 8192 }, "cost": { - "input": 1, - "output": 2, - "cache_read": 0.25 + "input": 0.6, + "output": 6 } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", + "us.anthropic.claude-sonnet-4-6": { + "id": "us.anthropic.claude-sonnet-4-6", + "name": "Claude Sonnet 4.6 (US)", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-01-15", - "last_updated": "2026-04-12", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 198000, + "context": 1000000, "output": 64000 }, "cost": { - "input": 3.75, - "output": 18.75, - "cache_read": 0.375, - "cache_write": 4.69 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openai-gpt-4o-2024-11-20": { - "id": "openai-gpt-4o-2024-11-20", - "name": "GPT-4o", - "family": "gpt", + "au.anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "au.anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5 (AU)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-28", - "last_updated": "2026-03-06", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 3.125, - "output": 12.5 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "llama-3.3-70b": { - "id": "llama-3.3-70b", - "name": "Llama 3.3 70B", + "meta.llama3-3-70b-instruct-v1:0": { + "id": "meta.llama3-3-70b-instruct-v1:0", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2023-12", - "release_date": "2025-04-06", - "last_updated": "2026-03-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] @@ -72460,666 +120373,715 @@ "output": 4096 }, "cost": { - "input": 0.7, - "output": 2.8 + "input": 0.72, + "output": 0.72 } }, - "kimi-k2-5": { - "id": "kimi-k2-5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": true, + "zai.glm-5": { + "id": "zai.glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-01-27", - "last_updated": "2026-04-30", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 65536 + "context": 202752, + "output": 101376 }, "cost": { - "input": 0.56, - "output": 3.5, - "cache_read": 0.22 + "input": 1, + "output": 3.2 } }, - "llama-3.2-3b": { - "id": "llama-3.2-3b", - "name": "Llama 3.2 3B", - "family": "llama", - "attachment": false, - "reasoning": false, + "us.anthropic.claude-opus-4-8": { + "id": "us.anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8 (US)", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-10-03", - "last_updated": "2026-03-12", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "arcee-trinity-large-thinking": { - "id": "arcee-trinity-large-thinking", - "name": "Trinity Large Thinking", - "family": "trinity", - "attachment": false, + "global.anthropic.claude-opus-4-7": { + "id": "global.anthropic.claude-opus-4-7", + "name": "Claude Opus 4.7 (Global)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] }, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-04", + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "us.anthropic.claude-fable-5": { + "id": "us.anthropic.claude-fable-5", + "name": "Claude Fable 5 (US)", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3125, - "output": 1.125, - "cache_read": 0.075 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "hermes-3-llama-3.1-405b": { - "id": "hermes-3-llama-3.1-405b", - "name": "Hermes 3 Llama 3.1 405b", - "family": "hermes", - "attachment": false, + "amazon.nova-lite-v1:0": { + "id": "amazon.nova-lite-v1:0", + "name": "Nova Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-lite", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-09-25", - "last_updated": "2026-03-12", + "knowledge": "2024-10", + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 300000, + "output": 8192 }, "cost": { - "input": 1.1, - "output": 3 + "input": 0.06, + "output": 0.24, + "cache_read": 0.015 } }, - "gemini-3-1-pro-preview": { - "id": "gemini-3-1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "us.anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "us.anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5 (US)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-19", - "last_updated": "2026-03-12", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.5, - "cache_write": 0.5, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - } - } - }, - "kimi-k2-6": { - "id": "kimi-k2-6", - "name": "Kimi K2.6", - "family": "kimi", + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 + } + }, + "xai.grok-4.3": { + "id": "xai.grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-30", + "release_date": "2026-04-17", + "last_updated": "2026-06-28", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 1000000, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/amazon-bedrock/mantle", + "api": "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.85, - "output": 4.655, - "cache_read": 0.22 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "claude-opus-4-6-fast": { - "id": "claude-opus-4-6-fast", - "name": "Claude Opus 4.6 Fast", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "mistral.voxtral-mini-3b-2507": { + "id": "mistral.voxtral-mini-3b-2507", + "name": "Voxtral Mini 3B 2507", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image"], + "input": ["audio", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 36, - "output": 180, - "cache_read": 3.6, - "cache_write": 45 + "input": 0.04, + "output": 0.04 } }, - "z-ai-glm-5-turbo": { - "id": "z-ai-glm-5-turbo", - "name": "GLM 5 Turbo", - "family": "glm", + "moonshot.kimi-k2-thinking": { + "id": "moonshot.kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-04-12", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 32768 + "context": 262143, + "output": 16000 }, "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24 + "input": 0.6, + "output": 2.5 } }, - "google-gemma-4-31b-it": { - "id": "google-gemma-4-31b-it", - "name": "Google Gemma 4 31B Instruct", - "family": "gemma", - "attachment": true, - "reasoning": true, + "meta.llama3-1-70b-instruct-v1:0": { + "id": "meta.llama3-1-70b-instruct-v1:0", + "name": "Llama 3.1 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-04-03", - "last_updated": "2026-04-12", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8192 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.175, - "output": 0.5 + "input": 0.72, + "output": 0.72 } - } - } - }, - "aihubmix": { - "id": "aihubmix", - "env": ["AIHUBMIX_API_KEY"], - "npm": "@aihubmix/ai-sdk-provider", - "name": "AIHubMix", - "doc": "https://docs.aihubmix.com", - "models": { - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "MiniMax-M2.7", - "family": "minimax", + }, + "us.deepseek.r1-v1:0": { + "id": "us.deepseek.r1-v1:0", + "name": "DeepSeek-R1 (US)", + "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.2958, - "output": 1.1832, - "cache_read": 0.05916 + "input": 1.35, + "output": 5.4 } }, - "coding-glm-5.1-free": { - "id": "coding-glm-5.1-free", - "name": "Coding GLM 5.1 (free)", - "family": "glm", - "attachment": false, + "global.anthropic.claude-sonnet-4-6": { + "id": "global.anthropic.claude-sonnet-4-6", + "name": "Claude Sonnet 4.6 (Global)", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-04-11", - "last_updated": "2026-04-11", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "gemini-3.1-pro-preview-customtools": { - "id": "gemini-3.1-pro-preview-customtools", - "name": "Gemini 3.1 Pro Preview Custom Tools", - "family": "gemini-pro", + "anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", + "moonshotai.kimi-k2.5": { + "id": "moonshotai.kimi-k2.5", "name": "Kimi K2.5", - "family": "kimi-k2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "interleaved": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": true, + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 0 + "context": 262143, + "output": 16000 }, "cost": { "input": 0.6, - "output": 3, - "cache_read": 0.105 + "output": 3 } }, - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM 5 Vision Turbo", - "family": "glm", + "au.anthropic.claude-opus-4-8": { + "id": "au.anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8 (AU)", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-05-09", - "last_updated": "2026-05-09", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 0.7042, - "output": 3.09848, - "cache_read": 0.169008 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "grok-4.3": { - "id": "grok-4.3", - "name": "Grok 4.3", - "family": "grok", - "attachment": true, - "reasoning": true, + "nvidia.nemotron-nano-12b-v2": { + "id": "nvidia.nemotron-nano-12b-v2", + "name": "NVIDIA Nemotron Nano 12B v2 VL BF16", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "family": "nemotron", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-05-01", - "last_updated": "2026-05-01", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 1000000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2, - "tiers": [ - { - "input": 2.5, - "output": 5, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 5, - "cache_read": 0.4 - } + "input": 0.2, + "output": 0.6 } }, - "coding-minimax-m2.7-highspeed": { - "id": "coding-minimax-m2.7-highspeed", - "name": "Coding MiniMax M2.7 Highspeed", - "family": "minimax", + "zai.glm-4.7-flash": { + "id": "zai.glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 13100 + "context": 200000, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.07, + "output": 0.4 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "meta.llama4-scout-17b-instruct-v1:0": { + "id": "meta.llama4-scout-17b-instruct-v1:0", + "name": "Llama 4 Scout 17B Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 3500000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.17, + "output": 0.66 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "au.anthropic.claude-sonnet-5": { + "id": "au.anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5 (AU)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { "input": 2, - "output": 12, - "cache_read": 0.2 + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "coding-glm-5.1": { - "id": "coding-glm-5.1", - "name": "Coding-GLM-5.1", - "family": "glm", + "qwen.qwen3-235b-a22b-2507-v1:0": { + "id": "qwen.qwen3-235b-a22b-2507-v1:0", + "name": "Qwen3 235B A22B 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-11", - "last_updated": "2026-04-11", + "knowledge": "2024-04", + "release_date": "2025-09-18", + "last_updated": "2025-09-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.06, - "output": 0.22 + "input": 0.22, + "output": 0.88 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "eu.anthropic.claude-haiku-4-5-20251001-v1:0": { + "id": "eu.anthropic.claude-haiku-4-5-20251001-v1:0", + "name": "Claude Haiku 4.5 (EU)", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 + "input": 1.1, + "output": 5.5, + "cache_read": 0.11, + "cache_write": 1.375 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "openai.gpt-oss-20b-1:0": { + "id": "openai.gpt-oss-20b-1:0", + "name": "gpt-oss-20b", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 128000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 + "input": 0.07, + "output": 0.3 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", + "jp.anthropic.claude-opus-4-8": { + "id": "jp.anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8 (JP)", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -73136,372 +121098,452 @@ "cache_write": 6.25 } }, - "deepseek-v4-flash-think": { - "id": "deepseek-v4-flash-think", - "name": "DeepSeek V4 Flash Think", - "family": "deepseek", - "attachment": false, + "anthropic.claude-opus-4-8": { + "id": "anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 128000 }, "cost": { - "input": 0.154, - "output": 0.308, - "cache_read": 0.0308 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "qwen.qwen3-coder-30b-a3b-v1:0": { + "id": "qwen.qwen3-coder-30b-a3b-v1:0", + "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-09-18", + "last_updated": "2025-09-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.15, + "output": 0.6 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "qwen.qwen3-next-80b-a3b": { + "id": "qwen.qwen3-next-80b-a3b", + "name": "Qwen/Qwen3-Next-80B-A3B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2025-09-18", + "last_updated": "2025-11-25", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.14, + "output": 1.4 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", + "openai.gpt-5.5": { + "id": "openai.gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 272000, "output": 128000 }, + "provider": { + "npm": "@ai-sdk/amazon-bedrock/mantle", + "api": "https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1", + "shape": "responses" + }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 5.5, + "output": 33, + "cache_read": 0.55 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen", + "au.anthropic.claude-sonnet-4-6": { + "id": "au.anthropic.claude-sonnet-4-6", + "name": "AU Anthropic Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-05-09", - "last_updated": "2026-05-09", + "knowledge": "2025-08", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 991000, - "output": 64000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.282, - "output": 1.692, - "cache_read": 0.0282, - "cache_write": 0.3525 + "input": 3.3, + "output": 16.5, + "cache_read": 0.33, + "cache_write": 4.125 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4-Mini", - "family": "gpt-mini", + "us.anthropic.claude-sonnet-5": { + "id": "us.anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5 (US)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "us.anthropic.claude-opus-4-5-20251101-v1:0": { + "id": "us.anthropic.claude-opus-4-5-20251101-v1:0", + "name": "Claude Opus 4.5 (US)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 128000 + "output": 64000 }, "cost": { - "input": 0.845, - "output": 3.38, - "cache_read": 0.183112 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", - "attachment": true, + "minimax.minimax-m2.5": { + "id": "minimax.minimax-m2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "structured_output": false, + "temperature": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 196608, + "output": 98304 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.275 + "input": 0.3, + "output": 1.2 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "family": "gpt-codex", + "eu.anthropic.claude-opus-4-8": { + "id": "eu.anthropic.claude-opus-4-8", + "name": "Claude Opus 4.8 (EU)", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 5.5, + "output": 27.5, + "cache_read": 0.55, + "cache_write": 6.875 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", + "eu.anthropic.claude-opus-4-7": { + "id": "eu.anthropic.claude-opus-4-7", + "name": "Claude Opus 4.7 (EU)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 2.499, - "cache_read": 0.03 + "input": 5.5, + "output": 27.5, + "cache_read": 0.55, + "cache_write": 6.875 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex mini", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "meta.llama3-1-8b-instruct-v1:0": { + "id": "meta.llama3-1-8b-instruct-v1:0", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.22, + "output": 0.22 } }, - "gemini-3.1-flash-lite": { - "id": "gemini-3.1-flash-lite", - "name": "Gemini 3.1 Flash Lite", - "family": "gemini-flash-lite", + "us.anthropic.claude-opus-4-1-20250805-v1:0": { + "id": "us.anthropic.claude-opus-4-1-20250805-v1:0", + "name": "Claude Opus 4.1 (US)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.25 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "meta.llama4-maverick-17b-instruct-v1:0": { + "id": "meta.llama4-maverick-17b-instruct-v1:0", + "name": "Llama 4 Maverick 17B Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-15", - "last_updated": "2025-11-15", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.24, + "output": 0.97 } }, - "claude-opus-4-6-think": { - "id": "claude-opus-4-6-think", - "name": "Claude Opus 4.6", + "global.anthropic.claude-opus-4-5-20251101-v1:0": { + "id": "global.anthropic.claude-opus-4-5-20251101-v1:0", + "name": "Claude Opus 4.5 (Global)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -73509,7 +121551,7 @@ "open_weights": false, "limit": { "context": 200000, - "output": 32000 + "output": 64000 }, "cost": { "input": 5, @@ -73518,123 +121560,156 @@ "cache_write": 6.25 } }, - "coding-minimax-m2.7-free": { - "id": "coding-minimax-m2.7-free", - "name": "Coding-MiniMax-M2.7-Free", - "family": "minimax", + "nvidia.nemotron-super-3-120b": { + "id": "nvidia.nemotron-super-3-120b", + "name": "NVIDIA Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 13100 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.65 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", - "attachment": false, + "eu.anthropic.claude-sonnet-5": { + "id": "eu.anthropic.claude-sonnet-5", + "name": "Claude Sonnet 5 (EU)", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 384000 + "output": 128000 }, "cost": { - "input": 0.154, - "output": 0.308, - "cache_read": 0.0308 + "input": 2.2, + "output": 11, + "cache_read": 0.22, + "cache_write": 2.75 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", - "attachment": true, + "nvidia.nemotron-nano-3-30b": { + "id": "nvidia.nemotron-nano-3-30b", + "name": "NVIDIA Nemotron Nano 3 30B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.95, - "output": 3.9995, - "cache_read": 0.160835 + "input": 0.06, + "output": 0.24 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + "eu.anthropic.claude-opus-4-5-20251101-v1:0": { + "id": "eu.anthropic.claude-opus-4-5-20251101-v1:0", + "name": "Claude Opus 4.5 (EU)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 5.5, + "output": 27.5, + "cache_read": 0.55, + "cache_write": 6.875 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", + "global.anthropic.claude-opus-4-6-v1": { + "id": "global.anthropic.claude-opus-4-6-v1", + "name": "Claude Opus 4.6 (Global)", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, "knowledge": "2025-05-31", "release_date": "2026-02-05", @@ -73655,338 +121730,441 @@ "cache_write": 6.25 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "us.anthropic.claude-sonnet-4-5-20250929-v1:0": { + "id": "us.anthropic.claude-sonnet-4-5-20250929-v1:0", + "name": "Claude Sonnet 4.5 (US)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "google.gemma-3-12b-it": { + "id": "google.gemma-3-12b-it", + "name": "Google Gemma 3 12B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2024-12", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 0.049999999999999996, + "output": 0.09999999999999999 + } + }, + "minimax.minimax-m2.1": { + "id": "minimax.minimax-m2.1", + "name": "MiniMax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.478, - "output": 0.956, - "cache_read": 0.004302 + "input": 0.3, + "output": 1.2 } }, - "claude-opus-4-7-think": { - "id": "claude-opus-4-7-think", - "name": "Claude Opus 4.7 Thinking", - "family": "claude-opus", + "eu.anthropic.claude-fable-5": { + "id": "eu.anthropic.claude-fable-5", + "name": "Claude Fable 5 (EU)", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 11, + "output": 55, + "cache_read": 1.1, + "cache_write": 13.75 } }, - "coding-minimax-m2.7": { - "id": "coding-minimax-m2.7", - "name": "Coding MiniMax M2.7", - "family": "minimax", + "deepseek.v3.2": { + "id": "deepseek.v3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2024-07", + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 13100 + "context": 163840, + "output": 81920 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.62, + "output": 1.85 } }, - "qwen3.6-max-preview": { - "id": "qwen3.6-max-preview", - "name": "Qwen3.6 Max Preview", - "family": "qwen", + "mistral.ministral-3-14b-instruct": { + "id": "mistral.ministral-3-14b-instruct", + "name": "Ministral 14B 3.0", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-05-09", - "last_updated": "2026-05-09", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 240000, - "output": 64000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 1.268, - "output": 7.608, - "cache_read": 0.1268, - "cache_write": 1.585 + "input": 0.2, + "output": 0.2 } - }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + } + } + }, + "umans-ai": { + "id": "umans-ai", + "env": ["UMANS_AI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.code.umans.ai/v1", + "name": "Umans AI", + "doc": "https://app.umans.ai/offers/code/docs/orgs", + "models": { + "umans-kimi-k2.7": { + "id": "umans-kimi-k2.7", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, + "context": 262144, "output": 32768 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": false, + "umans-glm-5.1": { + "id": "umans-glm-5.1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 1.4, + "output": 4.4, + "cache_read": 0.29 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "umans-coder": { + "id": "umans-coder", + "name": "Umans Coder", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "claude-sonnet-4-6-think": { - "id": "claude-sonnet-4-6-think", - "name": "Claude Sonnet 4.6 Think", - "family": "claude-sonnet", + "umans-flash": { + "id": "umans-flash", + "name": "Umans Flash", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.15, + "output": 1, + "cache_read": 0.05 } }, - "qwen3.6-flash": { - "id": "qwen3.6-flash", - "name": "Qwen3.6 Flash", - "family": "qwen", - "attachment": true, + "umans-glm-5.2": { + "id": "umans-glm-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 991000, - "output": 64000 + "context": 405504, + "output": 131072 }, "cost": { - "input": 0.169, - "output": 1.014, - "cache_read": 0.0169, - "cache_write": 0.21125 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } } } }, - "cerebras": { - "id": "cerebras", - "env": ["CEREBRAS_API_KEY"], - "npm": "@ai-sdk/cerebras", - "name": "Cerebras", - "doc": "https://inference-docs.cerebras.ai/models/overview", + "togetherai": { + "id": "togetherai", + "env": ["TOGETHER_API_KEY"], + "npm": "@ai-sdk/togetherai", + "name": "Together AI", + "doc": "https://docs.together.ai/docs/serverless-models", "models": { - "llama3.1-8b": { - "id": "llama3.1-8b", - "name": "Llama 3.1 8B", - "family": "llama", + "LiquidAI/LFM2-24B-A2B": { + "id": "LiquidAI/LFM2-24B-A2B", + "name": "LFM2-24B-A2B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "liquid", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-02-25", + "last_updated": "2026-02-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 8000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.03, + "output": 0.12 } }, - "qwen-3-235b-a22b-instruct-2507": { - "id": "qwen-3-235b-a22b-instruct-2507", - "name": "Qwen 3 235B Instruct", - "family": "qwen", + "meta-llama/Meta-Llama-3-8B-Instruct-Lite": { + "id": "meta-llama/Meta-Llama-3-8B-Instruct-Lite", + "name": "Meta Llama 3 8B Instruct Lite", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-22", - "last_updated": "2025-07-22", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131000, - "output": 32000 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 1.2 + "input": 0.14, + "output": 0.14 } }, - "zai-glm-4.7": { - "id": "zai-glm-4.7", - "name": "Z.AI GLM-4.7", + "meta-llama/Llama-3.3-70B-Instruct-Turbo": { + "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo", + "name": "Llama 3.3 70B", + "description": "Compact Llama instruction model for fast chat and local deployment", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-01-10", - "last_updated": "2026-01-10", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2026-07-02", "modalities": { "input": ["text"], "output": ["text"] @@ -73994,84 +122172,92 @@ "open_weights": true, "limit": { "context": 131072, - "output": 40000 + "output": 131072 }, "cost": { - "input": 2.25, - "output": 2.75, - "cache_read": 0, - "cache_write": 0 + "input": 1.04, + "output": 1.04 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "output": 131000 }, "cost": { - "input": 0.25, - "output": 0.69 + "input": 1.2, + "output": 4.5, + "cache_read": 0.2 } - } - } - }, - "lmstudio": { - "id": "lmstudio", - "env": ["LMSTUDIO_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "http://127.0.0.1:1234/v1", - "name": "LMStudio", - "doc": "https://lmstudio.ai/models", - "models": { - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + }, + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2026-01", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 2.8 } }, - "qwen/qwen3-coder-30b": { - "id": "qwen/qwen3-coder-30b", - "name": "Qwen3 Coder 30B", - "family": "qwen", + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Kimi coding model for software agents, refactors, and repository reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-06-14", + "last_updated": "2026-06-14", "modalities": { "input": ["text"], "output": ["text"] @@ -74079,174 +122265,192 @@ "open_weights": true, "limit": { "context": 262144, - "output": 65536 + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } }, - "qwen/qwen3-30b-a3b-2507": { - "id": "qwen/qwen3-30b-a3b-2507", - "name": "Qwen3 30B A3B 2507", - "family": "qwen", + "google/gemma-4-31B-it": { + "id": "google/gemma-4-31B-it", + "name": "Gemma 4 31B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 131072 + }, + "cost": { + "input": 0.39, + "output": 0.97 + } + }, + "google/gemma-3n-E4B-it": { + "id": "google/gemma-3n-E4B-it", + "name": "Gemma 3N E4B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.06, + "output": 0.12 } - } - } - }, - "lucidquery": { - "id": "lucidquery", - "env": ["LUCIDQUERY_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://lucidquery.com/api/v1", - "name": "LucidQuery AI", - "doc": "https://lucidquery.com/api/docs", - "models": { - "lucidnova-rf1-100b": { - "id": "lucidnova-rf1-100b", - "name": "LucidNova RF1 100B", - "family": "nova", - "attachment": true, - "reasoning": true, + }, + "Qwen/Qwen3.7-Max": { + "id": "Qwen/Qwen3.7-Max", + "name": "Qwen3.7 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2025-09-16", - "release_date": "2024-12-28", - "last_updated": "2025-09-10", + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-07-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 120000, - "output": 8000 + "context": 1000000, + "output": 500000 }, "cost": { - "input": 2, - "output": 5 + "input": 1.25, + "output": 3.75 } }, - "lucidquery-nexus-coder": { - "id": "lucidquery-nexus-coder", - "name": "LucidQuery Nexus Coder", - "family": "lucid", - "attachment": true, + "Qwen/Qwen3.6-Plus": { + "id": "Qwen/Qwen3.6-Plus", + "name": "Qwen3.6 Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-01", - "release_date": "2025-09-01", - "last_updated": "2025-09-01", + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 250000, - "output": 60000 + "context": 1000000, + "output": 500000 }, "cost": { - "input": 2, - "output": 5 + "input": 0.5, + "output": 3 } - } - } - }, - "moonshotai-cn": { - "id": "moonshotai-cn", - "env": ["MOONSHOT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.moonshot.cn/v1", - "name": "Moonshot AI (China)", - "doc": "https://platform.moonshot.cn/docs/api/chat", - "models": { - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + }, + "Qwen/Qwen3.5-397B-A17B": { + "id": "Qwen/Qwen3.5-397B-A17B", + "name": "Qwen3.5 397B A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "release_date": "2026-02-16", + "last_updated": "2026-06-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 130000 }, "cost": { "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "output": 3.6 } }, - "kimi-k2-0711-preview": { - "id": "kimi-k2-0711-preview", - "name": "Kimi K2 0711", - "family": "kimi", + "Qwen/Qwen3-Coder-Next-FP8": { + "id": "Qwen/Qwen3-Coder-Next-FP8", + "name": "Qwen3 Coder Next FP8", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-14", - "last_updated": "2025-07-14", + "knowledge": "2026-02-03", + "release_date": "2026-02-03", + "last_updated": "2026-02-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.5, + "output": 1.2 } }, - "kimi-k2-turbo-preview": { - "id": "kimi-k2-turbo-preview", - "name": "Kimi K2 Turbo", - "family": "kimi", + "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": { + "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Legacy model retained for compatibility with older integrations", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] @@ -74256,29 +122460,26 @@ "context": 262144, "output": 262144 }, + "status": "deprecated", "cost": { - "input": 2.4, - "output": 10, - "cache_read": 0.6 + "input": 2, + "output": 2 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", - "attachment": true, - "reasoning": true, + "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": { + "id": "Qwen/Qwen3-235B-A22B-Instruct-2507-tput", + "name": "Qwen3 235B A22B Instruct 2507 FP8", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-07", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, @@ -74287,803 +122488,969 @@ "output": 262144 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.2, + "output": 0.6 } }, - "kimi-k2-thinking-turbo": { - "id": "kimi-k2-thinking-turbo", - "name": "Kimi K2 Thinking Turbo", - "family": "kimi-thinking", + "Qwen/Qwen2.5-7B-Instruct-Turbo": { + "id": "Qwen/Qwen2.5-7B-Instruct-Turbo", + "name": "Qwen 2.5 7B Instruct Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.15, - "output": 8, - "cache_read": 0.15 + "input": 0.3, + "output": 0.3 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi-k2.5", - "attachment": false, + "Qwen/Qwen3.5-9B": { + "id": "Qwen/Qwen3.5-9B", + "name": "Qwen3.5 9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "temperature": true, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 65536 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.17, + "output": 0.25 } }, - "kimi-k2-0905-preview": { - "id": "kimi-k2-0905-preview", - "name": "Kimi K2 0905", - "family": "kimi", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2025-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.15, + "output": 0.6 } - } - } - }, - "azure-cognitive-services": { - "id": "azure-cognitive-services", - "env": ["AZURE_COGNITIVE_SERVICES_RESOURCE_NAME", "AZURE_COGNITIVE_SERVICES_API_KEY"], - "npm": "@ai-sdk/azure", - "name": "Azure Cognitive Services", - "doc": "https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models", - "models": { - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, + }, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-02-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 131072, + "output": 131072 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.05, + "output": 0.2 } }, - "claude-opus-4-1": { - "id": "claude-opus-4-1", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, + "pearl-ai/gemma-4-31b-it": { + "id": "pearl-ai/gemma-4-31b-it", + "name": "Pearl AI Gemma 4 31B Instruct", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 32000, "output": 32000 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" - }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.28, + "output": 0.86 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "nvidia/nemotron-3-ultra-550b-a55b": { + "id": "nvidia/nemotron-3-ultra-550b-a55b", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 512300, + "output": 512300 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "cost": { + "input": 0.6, + "output": 3.6, + "cache_read": 0.2 + } + }, + "deepcogito/cogito-v2-1-671b": { + "id": "deepcogito/cogito-v2-1-671b", + "name": "Cogito v2.1 671B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "cogito", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "temperature": true, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 163840, + "output": 163840 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 1.25, + "output": 1.25 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", - "shape": "completions" + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.95, - "output": 4 + "input": 1, + "output": 3.2 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-06-16", + "last_updated": "2026-06-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 262144, + "output": 164000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "knowledge": "2025-11", + "release_date": "2026-04-07", + "last_updated": "2026-07-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 202752, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.4, + "output": 4.4 } }, - "mai-ds-r1": { - "id": "mai-ds-r1", - "name": "MAI-DS-R1", - "family": "mai", + "deepseek-ai/DeepSeek-R1": { + "id": "deepseek-ai/DeepSeek-R1", + "name": "DeepSeek-R1", + "description": "Legacy model retained for compatibility with older integrations", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": false, "temperature": true, - "knowledge": "2024-06", + "knowledge": "2024-07", "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "last_updated": "2025-03-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 163839, + "output": 163839 }, + "status": "deprecated", "cost": { - "input": 1.35, - "output": 5.4 + "input": 3, + "output": 7 } }, - "llama-4-maverick-17b-128e-instruct-fp8": { - "id": "llama-4-maverick-17b-128e-instruct-fp8", - "name": "Llama 4 Maverick 17B 128E Instruct FP8", - "family": "llama", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-V3-1": { + "id": "deepseek-ai/DeepSeek-V3-1", + "name": "DeepSeek V3.1", + "description": "Legacy model retained for compatibility with older integrations", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2025-08", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 0.25, - "output": 1 + "input": 0.6, + "output": 1.7 } }, - "codestral-2501": { - "id": "codestral-2501", - "name": "Codestral 25.01", - "family": "codestral", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "DeepSeek V4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-24", + "last_updated": "2026-04-24", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 512000, + "output": 384000 + }, + "cost": { + "input": 1.74, + "output": 3.48, + "cache_read": 0.2 + } + }, + "deepseek-ai/DeepSeek-V3": { + "id": "deepseek-ai/DeepSeek-V3", + "name": "DeepSeek-V3", + "description": "Legacy model retained for compatibility with older integrations", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-03", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-07", + "release_date": "2024-12-26", + "last_updated": "2025-05-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 131072, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 0.3, - "output": 0.9 + "input": 1.25, + "output": 1.25 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 204800, + "output": 131072 }, + "status": "deprecated", "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "MiniMaxAI/MiniMax-M3": { + "id": "MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-12-02", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 524288, + "output": 250000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek-R1-0528", - "family": "deepseek-thinking", + "MiniMaxAI/MiniMax-M2.7": { + "id": "MiniMaxAI/MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 163840, - "output": 163840 + "context": 202752, + "output": 131072 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "gpt-3.5-turbo-instruct": { - "id": "gpt-3.5-turbo-instruct", - "name": "GPT-3.5 Turbo Instruct", - "family": "gpt", + "essentialai/Rnj-1-Instruct": { + "id": "essentialai/Rnj-1-Instruct", + "name": "Rnj-1 Instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "rnj", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-09-21", - "last_updated": "2023-09-21", + "knowledge": "2024-10", + "release_date": "2025-12-05", + "last_updated": "2025-12-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 4096, - "output": 4096 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.5, - "output": 2 + "input": 0.15, + "output": 0.15 } - }, - "mistral-medium-2505": { - "id": "mistral-medium-2505", - "name": "Mistral Medium 3", - "family": "mistral-medium", + } + } + }, + "frogbot": { + "id": "frogbot", + "env": ["FROGBOT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://app.frogbot.ai/api/v1", + "name": "FrogBot", + "doc": "https://docs.frogbot.ai", + "models": { + "minimax-m2-5": { + "id": "minimax-m2-5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "knowledge": "2024-09", + "release_date": "2025-01-15", + "last_updated": "2025-02-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 192000, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "phi-4-reasoning-plus": { - "id": "phi-4-reasoning-plus", - "name": "Phi-4-reasoning-plus", - "family": "phi", + "kimi-k2-6": { + "id": "kimi-k2-6", + "name": "Kimi-K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "release_date": "1970-01-01", + "last_updated": "1970-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 4096 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "cohere-embed-v3-english": { - "id": "cohere-embed-v3-english", - "name": "Embed v3 English", - "family": "cohere-embed", - "attachment": false, + "zai-glm-5-1": { + "id": "zai-glm-5-1", + "name": "Z.AI GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-11-07", - "last_updated": "2023-11-07", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2025-01-20", + "last_updated": "2025-02-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 512, - "output": 1024 + "context": 198000, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "gpt-4-32k": { - "id": "gpt-4-32k", - "name": "GPT-4 32K", - "family": "gpt", + "grok-code-fast-1": { + "id": "grok-code-fast-1", + "name": "Grok 4.1 Fast (Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-03-14", - "last_updated": "2023-03-14", + "knowledge": "2023-10", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 256000, + "output": 128000 }, "cost": { - "input": 60, - "output": 120 + "input": 0.2, + "output": 1.5, + "cache_read": 0.02 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { "input": 1.25, "output": 10, - "cache_read": 0.13 + "cache_read": 0.31 } }, - "phi-4": { - "id": "phi-4", - "name": "Phi-4", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2025-01", + "release_date": "2025-07-17", + "last_updated": "2025-07-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 0.3, + "output": 2.5, + "cache_read": 0.075 } }, - "gpt-3.5-turbo-0613": { - "id": "gpt-3.5-turbo-0613", - "name": "GPT-3.5 Turbo 0613", + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "family": "gpt", - "attachment": false, + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-06-13", - "last_updated": "2023-06-13", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, + "context": 128000, "output": 16384 }, "cost": { - "input": 3, - "output": 4 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "phi-3-medium-128k-instruct": { - "id": "phi-3-medium-128k-instruct", - "name": "Phi-3-medium-instruct (128k)", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, + "qwen-3-6-plus": { + "id": "qwen-3-6-plus", + "name": "Qwen 3.6 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "release_date": "2026-04-02", + "last_updated": "2026-04-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.5, + "output": 3, + "cache_read": 0.1 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek-V3.2", - "family": "deepseek", - "attachment": false, + "grok-4-3": { + "id": "grok-4-3", + "name": "Grok 4.3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-11", + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 0.58, - "output": 1.68 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "phi-3-small-128k-instruct": { - "id": "phi-3-small-128k-instruct", - "name": "Phi-3-small-instruct (128k)", - "family": "phi", - "attachment": false, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek v4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "knowledge": "2026-01", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 1.74, + "output": 3.48, + "cache_read": 0.14 } }, - "gpt-3.5-turbo-0301": { - "id": "gpt-3.5-turbo-0301", - "name": "GPT-3.5 Turbo 0301", - "family": "gpt", - "attachment": false, + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 + } + }, + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "Grok 4.1 Fast (Non-Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", + "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-03-01", - "last_updated": "2023-03-01", + "knowledge": "2025-11", + "release_date": "2025-11-25", + "last_updated": "2025-11-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 2000000, + "output": 128000 }, "cost": { - "input": 1.5, - "output": 2 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "phi-4-mini": { - "id": "phi-4-mini", - "name": "Phi-4-mini", - "family": "phi", - "attachment": false, + "minimax-m2-7": { + "id": "minimax-m2-7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2024-09", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 192000, + "output": 8192 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "GPT-5-Codex", - "family": "gpt-codex", - "attachment": false, + "gpt-5-3-codex": { + "id": "gpt-5-3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "knowledge": "2026-01-31", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -75094,4323 +123461,5221 @@ "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "meta-llama-3-8b-instruct": { - "id": "meta-llama-3-8b-instruct", - "name": "Meta-Llama-3-8B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "gpt-5-4-nano": { + "id": "gpt-5-4-nano", + "name": "GPT-5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.61 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "gpt-4": { - "id": "gpt-4", - "name": "GPT-4", - "family": "gpt", - "attachment": false, - "reasoning": false, + "gpt-5-4-mini": { + "id": "gpt-5-4-mini", + "name": "GPT-5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-03-14", - "last_updated": "2023-03-14", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 400000, + "output": 128000 }, "cost": { - "input": 60, - "output": 120 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "phi-4-mini-reasoning": { - "id": "phi-4-mini-reasoning", - "name": "Phi-4-mini-reasoning", - "family": "phi", + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi-K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "release_date": "1970-01-01", + "last_updated": "1970-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "meta-llama-3.1-70b-instruct": { - "id": "meta-llama-3.1-70b-instruct", - "name": "Meta-Llama-3.1-70B-Instruct", - "family": "llama", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "1970-01-01", + "last_updated": "1970-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, + "context": 131072, "output": 32768 }, "cost": { - "input": 2.68, - "output": 3.54 + "input": 0.15, + "output": 0.6 } }, - "phi-3-mini-4k-instruct": { - "id": "phi-3-mini-4k-instruct", - "name": "Phi-3-mini-instruct (4k)", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "gpt-5-5": { + "id": "gpt-5-5", + "name": "GPT-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 1024 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "deepseek-v3.1": { - "id": "deepseek-v3.1", - "name": "DeepSeek-V3.1", - "family": "deepseek", - "attachment": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.56, - "output": 1.68 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "text-embedding-3-small": { - "id": "text-embedding-3-small", - "name": "text-embedding-3-small", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 1536 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.02, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "gpt-3.5-turbo-1106": { - "id": "gpt-3.5-turbo-1106", - "name": "GPT-3.5 Turbo 1106", - "family": "gpt", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gemini-3-1-pro-preview": { + "id": "gemini-3-1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-11-06", - "last_updated": "2023-11-06", + "knowledge": "2026-01", + "release_date": "2026-02-18", + "last_updated": "2026-02-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 1, - "output": 2 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "model-router": { - "id": "model-router", - "name": "Model Router", - "family": "model-router", + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "Grok 4.1 Fast (Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "release_date": "2025-05-19", - "last_updated": "2025-11-18", + "temperature": true, + "knowledge": "2025-11", + "release_date": "2025-11-25", + "last_updated": "2025-11-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 2000000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "mistral-small-2503": { - "id": "mistral-small-2503", - "name": "Mistral Small 3.1", - "family": "mistral-small", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "o1": { - "id": "o1", - "name": "o1", - "family": "o", - "attachment": false, + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.5, + "output": 3, + "cache_read": 0.05 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "Grok 4 Fast (Reasoning)", - "family": "grok", - "attachment": true, + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "release_date": "1970-01-01", + "last_updated": "1970-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.07, + "output": 0.2 + } + } + } + }, + "openrouter": { + "id": "openrouter", + "env": ["OPENROUTER_API_KEY"], + "npm": "@openrouter/ai-sdk-provider", + "api": "https://openrouter.ai/api/v1", + "name": "OpenRouter", + "doc": "https://openrouter.ai/models", + "models": { + "inclusionai/ling-2.6-1t": { + "id": "inclusionai/ling-2.6-1t", + "name": "Ling-2.6-1T", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "family": "ling", + "attachment": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.075, + "output": 0.625, + "cache_read": 0.015 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, + "inclusionai/ring-2.6-1t": { + "id": "inclusionai/ring-2.6-1t", + "name": "Ring-2.6-1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "ring", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2026-05-08", + "last_updated": "2026-05-08", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.075, + "output": 0.625, + "cache_read": 0.015 + } + }, + "inclusionai/ling-2.6-flash": { + "id": "inclusionai/ling-2.6-flash", + "name": "Ling-2.6-flash", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "ling", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.01, + "output": 0.03, + "cache_read": 0.002 } }, - "cohere-embed-v3-multilingual": { - "id": "cohere-embed-v3-multilingual", - "name": "Embed v3 Multilingual", - "family": "cohere-embed", + "ibm-granite/granite-4.0-h-micro": { + "id": "ibm-granite/granite-4.0-h-micro", + "name": "Granite 4.0 Micro", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "granite", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2023-11-07", - "last_updated": "2023-11-07", + "structured_output": false, + "temperature": true, + "release_date": "2025-10-20", + "last_updated": "2025-10-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 512, - "output": 1024 + "context": 131000, + "output": 131000 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.017, + "output": 0.112 } }, - "o1-preview": { - "id": "o1-preview", - "name": "o1-preview", - "family": "o", + "ibm-granite/granite-4.1-8b": { + "id": "ibm-granite/granite-4.1-8b", + "name": "Granite 4.1 8B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "granite", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "structured_output": true, + "temperature": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 16.5, - "output": 66, - "cache_read": 8.25 + "input": 0.05, + "output": 0.1, + "cache_read": 0.05 } }, - "gpt-3.5-turbo-0125": { - "id": "gpt-3.5-turbo-0125", - "name": "GPT-3.5 Turbo 0125", - "family": "gpt", + "meta-llama/llama-3.1-8b-instruct": { + "id": "meta-llama/llama-3.1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2021-08", - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "knowledge": "2023-12-31", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16384, + "context": 131072, "output": 16384 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.02, + "output": 0.03 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex Mini", - "family": "gpt-codex", + "meta-llama/llama-3.1-70b-instruct": { + "id": "meta-llama/llama-3.1-70b-instruct", + "name": "Llama 3.1 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.4, + "output": 0.4 } }, - "cohere-embed-v-4-0": { - "id": "cohere-embed-v-4-0", - "name": "Embed v4", - "family": "cohere-embed", - "attachment": true, + "meta-llama/llama-3.2-1b-instruct": { + "id": "meta-llama/llama-3.2-1b-instruct", + "name": "Llama 3.2 1B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "structured_output": false, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 1536 + "context": 131072, + "output": 60000 }, "cost": { - "input": 0.12, - "output": 0 + "input": 0.027, + "output": 0.201 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": false, - "reasoning": true, + "meta-llama/llama-4-maverick": { + "id": "meta-llama/llama-4-maverick", + "name": "Llama 4 Maverick", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.2, + "output": 0.8 } }, - "gpt-4-turbo-vision": { - "id": "gpt-4-turbo-vision", - "name": "GPT-4 Turbo Vision", - "family": "gpt", + "meta-llama/llama-3.2-11b-vision-instruct": { + "id": "meta-llama/llama-3.2-11b-vision-instruct", + "name": "Llama 3.2 11B Vision Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "knowledge": "2023-12-31", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, - "limit": { - "context": 128000, - "output": 4096 - }, - "cost": { - "input": 10, - "output": 30 - } - }, - "gpt-5.1-chat": { - "id": "gpt-5.1-chat", - "name": "GPT-5.1 Chat", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", - "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] - }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, + "context": 131072, "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.345, + "output": 0.345 } }, - "meta-llama-3.1-405b-instruct": { - "id": "meta-llama-3.1-405b-instruct", - "name": "Meta-Llama-3.1-405B-Instruct", + "meta-llama/llama-3.3-70b-instruct:free": { + "id": "meta-llama/llama-3.3-70b-instruct:free", + "name": "Llama 3.3 70B Instruct (free)", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 5.33, - "output": 16 + "input": 0, + "output": 0 } }, - "llama-3.2-11b-vision-instruct": { - "id": "llama-3.2-11b-vision-instruct", - "name": "Llama-3.2-11B-Vision-Instruct", + "meta-llama/llama-3.3-70b-instruct": { + "id": "meta-llama/llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", "family": "llama", - "attachment": true, + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.37, - "output": 0.37 + "input": 0.1, + "output": 0.32 } }, - "cohere-command-a": { - "id": "cohere-command-a", - "name": "Command A", - "family": "command-a", + "meta-llama/llama-3.2-3b-instruct:free": { + "id": "meta-llama/llama-3.2-3b-instruct:free", + "name": "Llama 3.2 3B Instruct (free)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "knowledge": "2023-12-31", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0, + "output": 0 } }, - "mistral-large-2411": { - "id": "mistral-large-2411", - "name": "Mistral Large 24.11", - "family": "mistral-large", - "attachment": false, + "meta-llama/llama-guard-4-12b": { + "id": "meta-llama/llama-guard-4-12b", + "name": "Llama Guard 4 12B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "llama", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "knowledge": "2024-08-31", + "release_date": "2025-04-30", + "last_updated": "2025-04-30", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 163840, + "output": 16384 }, "cost": { - "input": 2, - "output": 6 + "input": 0.18, + "output": 0.18 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", + "meta-llama/llama-4-scout": { + "id": "meta-llama/llama-4-scout", + "name": "Llama 4 Scout", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 10000000, + "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.125 + "input": 0.1, + "output": 0.3 } }, - "deepseek-v3.2-speciale": { - "id": "deepseek-v3.2-speciale", - "name": "DeepSeek-V3.2-Speciale", - "family": "deepseek", + "meta-llama/llama-3.2-3b-instruct": { + "id": "meta-llama/llama-3.2-3b-instruct", + "name": "Llama 3.2 3B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2023-12-31", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.58, - "output": 1.68 + "input": 0.05, + "output": 0.33 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", - "attachment": false, + "~anthropic/claude-haiku-latest": { + "id": "~anthropic/claude-haiku-latest", + "name": "Anthropic Claude Haiku Latest", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "llama-3.2-90b-vision-instruct": { - "id": "llama-3.2-90b-vision-instruct", - "name": "Llama-3.2-90B-Vision-Instruct", - "family": "llama", + "~anthropic/claude-fable-latest": { + "id": "~anthropic/claude-fable-latest", + "name": "Claude Fable Latest", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "structured_output": true, + "temperature": false, + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2.04, - "output": 2.04 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "text-embedding-ada-002": { - "id": "text-embedding-ada-002", - "name": "text-embedding-ada-002", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2022-12-15", - "last_updated": "2022-12-15", + "~anthropic/claude-sonnet-latest": { + "id": "~anthropic/claude-sonnet-latest", + "name": "Anthropic Claude Sonnet Latest", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", - "attachment": false, + "~anthropic/claude-opus-latest": { + "id": "~anthropic/claude-opus-latest", + "name": "Claude Opus Latest", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "phi-3-small-8k-instruct": { - "id": "phi-3-small-8k-instruct", - "name": "Phi-3-small-instruct (8k)", - "family": "phi", + "moonshotai/kimi-k2": { + "id": "moonshotai/kimi-k2", + "name": "Kimi K2 0711", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "knowledge": "2024-12-31", + "release_date": "2025-07-11", + "last_updated": "2025-07-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 2048 + "context": 131072, + "output": 100352 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.57, + "output": 2.3 } }, - "meta-llama-3-70b-instruct": { - "id": "meta-llama-3-70b-instruct", - "name": "Meta-Llama-3-70B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 2048 + "context": 262144, + "output": 262144 }, "cost": { - "input": 2.68, - "output": 3.54 + "input": 0.72, + "output": 3.5, + "cache_read": 0.15 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Thinking Kimi model for slower research passes, planning, and hard technical questions", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 272000, - "output": 128000 + "context": 262144, + "output": 100352 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.01 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 272000, - "output": 128000 + "context": 262144, + "output": 256000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.03 + "input": 0.375, + "output": 2.025, + "cache_read": 0.203 } }, - "phi-4-reasoning": { - "id": "phi-4-reasoning", - "name": "Phi-4-reasoning", - "family": "phi", - "attachment": false, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 0.66, + "output": 3.41, + "cache_read": 0.15 } }, - "phi-3-mini-128k-instruct": { - "id": "phi-3-mini-128k-instruct", - "name": "Phi-3-mini-instruct (128k)", - "family": "phi", + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "knowledge": "2024-12-31", + "release_date": "2025-09-04", + "last_updated": "2025-09-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 100352 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 0.6, + "output": 2.5 } }, - "text-embedding-3-large": { - "id": "text-embedding-3-large", - "name": "text-embedding-3-large", - "family": "text-embedding", - "attachment": false, - "reasoning": false, + "baidu/ernie-4.5-vl-424b-a47b": { + "id": "baidu/ernie-4.5-vl-424b-a47b", + "name": "ERNIE 4.5 VL 424B A47B ", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "ernie", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "structured_output": false, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8191, - "output": 3072 + "context": 131072, + "output": 16000 }, "cost": { - "input": 0.13, - "output": 0 + "input": 0.42, + "output": 1.25 } }, - "o1-mini": { - "id": "o1-mini", - "name": "o1-mini", - "family": "o-mini", - "attachment": false, + "perceptron/perceptron-mk1": { + "id": "perceptron/perceptron-mk1", + "name": "Perceptron Mk1", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 32768, + "output": 8192 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.15, + "output": 1.5 } }, - "phi-3.5-moe-instruct": { - "id": "phi-3.5-moe-instruct", - "name": "Phi-3.5-MoE-instruct", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.16, - "output": 0.64 + "input": 0.25, + "output": 1.5, + "reasoning": 1.5, + "cache_read": 0.025, + "cache_write": 0.083333 } }, - "gpt-5-chat": { - "id": "gpt-5-chat", - "name": "GPT-5 Chat", - "family": "gpt-codex", + "google/gemini-3.1-flash-image": { + "id": "google/gemini-3.1-flash-image", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], "tool_call": false, - "temperature": false, - "knowledge": "2024-10-24", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.5, + "output": 3 } }, - "deepseek-v3-0324": { - "id": "deepseek-v3-0324", - "name": "DeepSeek-V3-0324", - "family": "deepseek", + "google/gemma-3n-e4b-it": { + "id": "google/gemma-3n-e4b-it", + "name": "Gemma 3n 4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "knowledge": "2024-08-31", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.14, - "output": 4.56 + "input": 0.06, + "output": 0.12 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "google/gemma-4-26b-a4b-it:free": { + "id": "google/gemma-4-26b-a4b-it:free", + "name": "Gemma 4 26B A4B (free)", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, + "context": 262144, "output": 32768 }, "cost": { - "input": 0.71, - "output": 0.71 + "input": 0, + "output": 0 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "interleaved": true, "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", - "shape": "completions" + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 3 + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "meta-llama-3.1-8b-instruct": { - "id": "meta-llama-3.1-8b-instruct", - "name": "Meta-Llama-3.1-8B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1048576, + "output": 65535 }, "cost": { "input": 0.3, - "output": 0.61 + "output": 2.5, + "reasoning": 2.5, + "cache_read": 0.03, + "cache_write": 0.083333 } }, - "ministral-3b": { - "id": "ministral-3b", - "name": "Ministral 3B", - "family": "ministral", - "attachment": false, - "reasoning": false, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 1.5, + "output": 9, + "reasoning": 9, + "cache_read": 0.15, + "cache_write": 0.083333 } }, - "phi-3-medium-4k-instruct": { - "id": "phi-3-medium-4k-instruct", - "name": "Phi-3-medium-instruct (4k)", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 4096, - "output": 1024 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.12, + "output": 0.35, + "cache_read": 0.09 } }, - "llama-4-scout-17b-16e-instruct": { - "id": "llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", + "google/lyria-3-clip-preview": { + "id": "google/lyria-3-clip-preview", + "name": "Lyria 3 Clip Preview", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "lyria", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2026-03-30", + "last_updated": "2026-03-30", "modalities": { "input": ["text", "image"], + "output": ["text", "audio"] + }, + "open_weights": false, + "limit": { + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "google/gemini-3.1-pro-preview-customtools": { + "id": "google/gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048756, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.78 + "input": 2, + "output": 12, + "reasoning": 12, + "cache_read": 0.2, + "cache_write": 0.375, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "phi-3.5-mini-instruct": { - "id": "phi-3.5-mini-instruct", - "name": "Phi-3.5-mini-instruct", - "family": "phi", - "attachment": false, - "reasoning": false, + "google/gemini-3.1-flash-lite-image": { + "id": "google/gemini-3.1-flash-lite-image", + "name": "Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2025-01-01", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["image", "text"], + "output": ["image", "text"] + }, + "open_weights": false, + "limit": { + "context": 65536, + "output": 66000 + }, + "cost": { + "input": 0.25, + "output": 1.5 + } + }, + "google/gemini-3-pro-image-preview": { + "id": "google/gemini-3-pro-image-preview", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "knowledge": "2025-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 65536, + "output": 32768 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 2, + "output": 12, + "reasoning": 12, + "cache_read": 0.2, + "cache_write": 0.375 } }, - "phi-4-multimodal": { - "id": "phi-4-multimodal", - "name": "Phi-4-multimodal", - "family": "phi", + "google/gemini-2.5-flash-image": { + "id": "google/gemini-2.5-flash-image", + "name": "Nano Banana", + "description": "Nano Banana image model for fast generation, edits, and character-consistent assets", + "family": "gemini-flash", "attachment": true, "reasoning": false, "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "knowledge": "2024-06", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.08, - "output": 0.32, - "input_audio": 4 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "cache_write": 0.083333 } }, - "codex-mini": { - "id": "codex-mini", - "name": "Codex Mini", - "family": "gpt-codex-mini", + "google/gemini-2.5-flash-lite": { + "id": "google/gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-05-16", - "last_updated": "2025-05-16", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 1.5, - "output": 6, - "cache_read": 0.375 + "input": 0.1, + "output": 0.4, + "reasoning": 0.4, + "cache_read": 0.01, + "cache_write": 0.083333 } }, - "gpt-5.2-chat": { - "id": "gpt-5.2-chat", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", + "google/gemini-3.1-flash-image-preview": { + "id": "google/gemini-3.1-flash-image-preview", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["image", "text"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.5, + "output": 3 } }, - "mistral-nemo": { - "id": "mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, + "google/gemini-2.5-pro-preview-05-06": { + "id": "google/gemini-2.5-pro-preview-05-06", + "name": "Gemini 2.5 Pro Preview 05-06", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2025-01-31", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf", "audio", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 2, + "output": 12, + "reasoning": 12, + "cache_read": 0.2, + "cache_write": 0.375, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.06, + "output": 0.33 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", + "google/gemini-2.5-pro-preview": { + "id": "google/gemini-2.5-pro-preview", + "name": "Gemini 2.5 Pro Preview 06-05", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01-31", + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "image"], + "input": ["pdf", "image", "text", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 30, - "output": 180, - "context_over_200k": { - "input": 60, - "output": 270 - }, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 1.25, + "output": 10, + "reasoning": 10, + "cache_read": 0.125, + "cache_write": 0.375 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.5, + "output": 3, + "reasoning": 3, + "cache_read": 0.05, + "cache_write": 0.083333 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", + "google/gemma-3-12b-it": { + "id": "google/gemma-3-12b-it", + "name": "Gemma 3 12B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "knowledge": "2024-08-31", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.05, + "output": 0.15 } }, - "grok-3": { - "id": "grok-3", - "name": "Grok 3", - "family": "grok", - "attachment": false, + "google/gemma-3-4b-it": { + "id": "google/gemma-3-4b-it", + "name": "Gemma 3 4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2024-08-31", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 8192 + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.05, + "output": 0.1 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", + "google/gemma-3-27b-it": { + "id": "google/gemma-3-27b-it", + "name": "Gemma 3 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2024-08-31", + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.08, + "output": 0.16 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "google/lyria-3-pro-preview": { + "id": "google/lyria-3-pro-preview", + "name": "Lyria 3 Pro Preview", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "lyria", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-03-30", + "last_updated": "2026-03-30", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0, + "output": 0 } }, - "cohere-command-r-plus-08-2024": { - "id": "cohere-command-r-plus-08-2024", - "name": "Command R+", - "family": "command-r", - "attachment": false, - "reasoning": false, + "google/gemma-4-31b-it:free": { + "id": "google/gemma-4-31b-it:free", + "name": "Gemma 4 31B (free)", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 262144, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0, + "output": 0 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, + "google/gemma-2-27b-it": { + "id": "google/gemma-2-27b-it", + "name": "Gemma 2 27B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "knowledge": "2024-06-30", + "release_date": "2024-07-13", + "last_updated": "2024-07-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 8192, + "output": 2048 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.65, + "output": 0.65 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", + "google/gemini-3-pro-image": { + "id": "google/gemini-3-pro-image", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 272000 + "context": 65536, + "output": 32768 }, "cost": { - "input": 15, - "output": 120 + "input": 2, + "output": 12, + "reasoning": 12, + "cache_read": 0.2, + "cache_write": 0.375 } }, - "o3": { - "id": "o3", - "name": "o3", - "family": "o", + "google/gemini-3.1-flash-lite-preview": { + "id": "google/gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.25, + "output": 1.5, + "reasoning": 1.5, + "cache_read": 0.025, + "cache_write": 0.083333 } }, - "grok-3-mini": { - "id": "grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", + "liquid/lfm-2.5-1.2b-thinking:free": { + "id": "liquid/lfm-2.5-1.2b-thinking:free", + "name": "LFM2.5-1.2B-Thinking (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "liquid", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2025-06", + "release_date": "2026-01-20", + "last_updated": "2026-01-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 0.5, - "reasoning": 0.5, - "cache_read": 0.075 + "input": 0, + "output": 0 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "GPT-4.1 nano", - "family": "gpt-nano", - "attachment": true, + "liquid/lfm-2.5-1.2b-instruct:free": { + "id": "liquid/lfm-2.5-1.2b-instruct:free", + "name": "LFM2.5-1.2B-Instruct (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "liquid", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-06", + "release_date": "2026-01-20", + "last_updated": "2026-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, + "context": 32768, "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.03 + "input": 0, + "output": 0 } }, - "grok-4": { - "id": "grok-4", - "name": "Grok 4", + "x-ai/grok-4.20": { + "id": "x-ai/grok-4.20", + "name": "Grok 4.20", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "family": "grok", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "knowledge": "2025-09-01", + "release_date": "2026-03-31", + "last_updated": "2026-03-31", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 3, - "output": 15, - "reasoning": 15, - "cache_read": 0.75 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o-mini", - "attachment": false, + "x-ai/grok-4.3": { + "id": "x-ai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "structured_output": true, + "temperature": true, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "output": 1000000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", + "x-ai/grok-4.5": { + "id": "x-ai/grok-4.5", + "name": "Grok 4.5", + "description": "xAI's latest Grok for chat, coding, agentic tools, and lower hallucination risk", "family": "grok", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 500000, + "output": 500000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 2, + "output": 6, + "cache_read": 0.5, + "tiers": [ + { + "input": 4, + "output": 12, + "cache_read": 1, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 12, + "cache_read": 1 + } } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", + "x-ai/grok-4.20-multi-agent": { + "id": "x-ai/grok-4.20-multi-agent", + "name": "Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "high"] + } + ], + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2025-09-01", + "release_date": "2026-03-31", + "last_updated": "2026-03-31", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "cohere-command-r-08-2024": { - "id": "cohere-command-r-08-2024", - "name": "Command R", - "family": "command-r", - "attachment": false, - "reasoning": false, + "x-ai/grok-build-0.1": { + "id": "x-ai/grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 1, + "output": 2, + "cache_read": 0.2 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", + "~google/gemini-pro-latest": { + "id": "~google/gemini-pro-latest", + "name": "Google Gemini Pro Latest", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image"], + "input": ["audio", "pdf", "image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 2, + "output": 12, + "reasoning": 12, + "cache_read": 0.2, + "cache_write": 0.375 } }, - "gpt-4-turbo": { - "id": "gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", + "~google/gemini-flash-latest": { + "id": "~google/gemini-flash-latest", + "name": "Google Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "knowledge": "2025-01-01", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "pdf", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 10, - "output": 30 + "input": 1.5, + "output": 9, + "reasoning": 9, + "cache_read": 0.15, + "cache_write": 0.083333 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, + "microsoft/phi-4": { + "id": "microsoft/phi-4", + "name": "Phi 4", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", + "attachment": false, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "temperature": true, + "knowledge": "2024-06-30", + "release_date": "2025-01-10", + "last_updated": "2025-01-10", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 16384, + "output": 16384 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.07, + "output": 0.14 } - } - } - }, - "abliteration-ai": { - "id": "abliteration-ai", - "env": ["ABLIT_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.abliteration.ai/v1", - "name": "abliteration.ai", - "doc": "https://docs.abliteration.ai/models", - "models": { - "abliterated-model": { - "id": "abliterated-model", - "name": "Abliterated Model", - "attachment": true, + }, + "microsoft/wizardlm-2-8x22b": { + "id": "microsoft/wizardlm-2-8x22b", + "name": "WizardLM-2 8x22B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, "temperature": true, - "release_date": "2026-01-06", - "last_updated": "2026-01-06", + "knowledge": "2024-04-30", + "release_date": "2024-04-16", + "last_updated": "2024-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 150000, - "input": 150000, - "output": 8192 + "context": 65536, + "output": 8000 }, "cost": { - "input": 3, - "output": 3 + "input": 0.62, + "output": 0.62 } - } - } - }, - "wafer.ai": { - "id": "wafer.ai", - "env": ["WAFER_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://pass.wafer.ai/v1", - "name": "Wafer", - "doc": "https://docs.wafer.ai/wafer-pass", - "models": { - "Qwen3.5-397B-A17B": { - "id": "Qwen3.5-397B-A17B", - "name": "Qwen3.5 397B A17B", - "family": "qwen", - "attachment": true, + }, + "poolside/laguna-xs-2.1:free": { + "id": "poolside/laguna-xs-2.1:free", + "name": "Laguna XS 2.1 (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "laguna", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "release_date": "2026-07-02", + "last_updated": "2026-07-02", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 65536 + "output": 32768 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "GLM-5.1": { - "id": "GLM-5.1", - "name": "GLM-5.1", - "family": "glm", + "poolside/laguna-m.1": { + "id": "poolside/laguna-m.1", + "name": "Laguna M.1", + "description": "Poolside's flagship agentic coding model for long-horizon work", + "family": "laguna", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "release_date": "2026-04-28", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.2, + "output": 0.4, + "cache_read": 0.1 } - } - } - }, - "cohere": { - "id": "cohere", - "env": ["COHERE_API_KEY"], - "npm": "@ai-sdk/cohere", - "name": "Cohere", - "doc": "https://docs.cohere.com/docs/models", - "models": { - "command-a-reasoning-08-2025": { - "id": "command-a-reasoning-08-2025", - "name": "Command A Reasoning", - "family": "command-a", + }, + "poolside/laguna-xs-2.1": { + "id": "poolside/laguna-xs-2.1", + "name": "Laguna XS 2.1", + "description": "Agentic coding model from Poolside in the XS size class for local deployment", + "family": "laguna", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "release_date": "2026-07-02", + "last_updated": "2026-07-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.06, + "output": 0.12, + "cache_read": 0.03 } }, - "command-r7b-12-2024": { - "id": "command-r7b-12-2024", - "name": "Command R7B", - "family": "command-r", + "poolside/laguna-m.1:free": { + "id": "poolside/laguna-m.1:free", + "name": "Laguna M.1 (free)", + "description": "Free provider route for experiments, demos, and cost-sensitive chat workloads", + "family": "laguna", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": false, + "temperature": true, + "release_date": "2026-04-28", + "last_updated": "2026-06-13", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 32768 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "writer/palmyra-x5": { + "id": "writer/palmyra-x5", + "name": "Palmyra X5", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "palmyra", "attachment": false, "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2026-01-21", + "last_updated": "2026-01-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1040000, + "output": 8192 + }, + "cost": { + "input": 0.6, + "output": 6 + } + }, + "z-ai/glm-4.7": { + "id": "z-ai/glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-02-27", - "last_updated": "2024-02-27", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0.0375, - "output": 0.15 + "input": 0.4, + "output": 1.75, + "cache_read": 0.08 } }, - "c4ai-aya-vision-8b": { - "id": "c4ai-aya-vision-8b", - "name": "Aya Vision 8B", + "z-ai/glm-4.5v": { + "id": "z-ai/glm-4.5v", + "name": "GLM-4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-03-04", - "last_updated": "2025-05-14", + "knowledge": "2025-04", + "release_date": "2025-08-11", + "last_updated": "2025-08-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4000 + "context": 65536, + "output": 16384 + }, + "cost": { + "input": 0.6, + "output": 1.8, + "cache_read": 0.11 } }, - "command-r-plus-08-2024": { - "id": "command-r-plus-08-2024", - "name": "Command R+", - "family": "command-r", + "z-ai/glm-4.5": { + "id": "z-ai/glm-4.5", + "name": "GLM-4.5", + "description": "Hybrid-reasoning GLM release that made the 4.5 line broadly useful", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 131072, + "output": 98304 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "c4ai-aya-expanse-8b": { - "id": "c4ai-aya-expanse-8b", - "name": "Aya Expanse 8B", + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-10-24", - "last_updated": "2024-10-24", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "output": 4000 + "context": 202752, + "output": 128000 + }, + "cost": { + "input": 0.966, + "output": 3.036, + "cache_read": 0.1794 } }, - "command-r7b-arabic-02-2025": { - "id": "command-r7b-arabic-02-2025", - "name": "Command R7B Arabic", - "family": "command-r", + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "GLM-4.6", + "description": "Late GLM-4 workhorse for coding agents, reasoning, and structured tasks", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-02-27", - "last_updated": "2025-02-27", + "knowledge": "2025-04", + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 200000, + "output": 16384 }, "cost": { - "input": 0.0375, - "output": 0.15 + "input": 0.43, + "output": 1.75, + "cache_read": 0.08 } }, - "command-a-vision-07-2025": { - "id": "command-a-vision-07-2025", - "name": "Command A Vision", - "family": "command-a", + "z-ai/glm-5.2": { + "id": "z-ai/glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.42, + "output": 1.32, + "cache_read": 0.078 } }, - "c4ai-aya-vision-32b": { - "id": "c4ai-aya-vision-32b", - "name": "Aya Vision 32B", + "z-ai/glm-4.6v": { + "id": "z-ai/glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-03-04", - "last_updated": "2025-05-14", + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4000 + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.3, + "output": 0.9, + "cache_read": 0.055 } }, - "command-a-translate-08-2025": { - "id": "command-a-translate-08-2025", - "name": "Command A Translate", - "family": "command-a", - "attachment": false, - "reasoning": false, + "z-ai/glm-5v-turbo": { + "id": "z-ai/glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "Fast GLM vision model for screenshots, documents, and multimodal agent tasks", + "family": "glm", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8000, - "output": 8000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "command-r-08-2024": { - "id": "command-r-08-2024", - "name": "Command R", - "family": "command-r", + "z-ai/glm-4.5-air": { + "id": "z-ai/glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Lighter GLM-4.5 variant for fast coding assistance and cheaper agents", + "family": "glm-air", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.13, + "output": 0.85, + "cache_read": 0.025 } }, - "c4ai-aya-expanse-32b": { - "id": "c4ai-aya-expanse-32b", - "name": "Aya Expanse 32B", + "z-ai/glm-4.7-flash": { + "id": "z-ai/glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, "temperature": true, - "release_date": "2024-10-24", - "last_updated": "2024-10-24", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 202752, + "output": 16384 + }, + "cost": { + "input": 0.06, + "output": 0.4, + "cache_read": 0.01 } }, - "command-a-03-2025": { - "id": "command-a-03-2025", - "name": "Command A", - "family": "command-a", + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 8000 + "context": 202752, + "output": 128000 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.6, + "output": 1.92, + "cache_read": 0.12 } - } - } - }, - "cloudferro-sherlock": { - "id": "cloudferro-sherlock", - "env": ["CLOUDFERRO_SHERLOCK_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api-sherlock.cloudferro.com/openai/v1/", - "name": "CloudFerro Sherlock", - "doc": "https://docs.sherlock.cloudferro.com/", - "models": { - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + }, + "z-ai/glm-5-turbo": { + "id": "z-ai/glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Faster GLM-5 lane for coding agents that need lower latency", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": false, "temperature": true, - "knowledge": "2024-10-09", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 70000, - "output": 70000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 2.92, - "output": 2.92 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "OpenAI GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, + "openai/gpt-4o-mini-2024-07-18": { + "id": "openai/gpt-4o-mini-2024-07-18", + "name": "GPT-4o-mini (2024-07-18)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "o-mini", + "attachment": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "knowledge": "2023-10-31", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2.92, - "output": 2.92 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "speakleash/Bielik-11B-v3.0-Instruct": { - "id": "speakleash/Bielik-11B-v3.0-Instruct", - "name": "Bielik 11B v3.0 Instruct", + "openai/gpt-oss-safeguard-20b": { + "id": "openai/gpt-oss-safeguard-20b", + "name": "gpt-oss-safeguard-20b", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 32000 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.67, - "output": 0.67 + "input": 0.075, + "output": 0.3, + "cache_read": 0.0375 } }, - "speakleash/Bielik-11B-v2.6-Instruct": { - "id": "speakleash/Bielik-11B-v2.6-Instruct", - "name": "Bielik 11B v2.6 Instruct", + "openai/gpt-3.5-turbo-instruct": { + "id": "openai/gpt-3.5-turbo-instruct", + "name": "GPT-3.5 Turbo Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "knowledge": "2021-09-30", + "release_date": "2023-09-28", + "last_updated": "2023-09-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 32000 + "context": 4095, + "output": 4096 }, "cost": { - "input": 0.67, - "output": 0.67 + "input": 1.5, + "output": 2 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, + "openai/gpt-5.2-chat": { + "id": "openai/gpt-5.2-chat", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-10", + "last_updated": "2025-12-10", "modalities": { - "input": ["text"], + "input": ["pdf", "image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196000, - "input": 180000, - "output": 16000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } - } - } - }, - "kuae-cloud-coding-plan": { - "id": "kuae-cloud-coding-plan", - "env": ["KUAE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://coding-plan-endpoint.kuaecloud.net/v1", - "name": "KUAE Cloud Coding Plan", - "doc": "https://docs.mthreads.com/kuaecloud/kuaecloud-doc-online/coding_plan/", - "models": { - "GLM-4.7": { - "id": "GLM-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + }, + "openai/gpt-5.6-luna-pro": { + "id": "openai/gpt-5.6-luna-pro", + "name": "GPT-5.6 Luna Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25 } - } - } - }, - "xai": { - "id": "xai", - "env": ["XAI_API_KEY"], - "npm": "@ai-sdk/xai", - "name": "xAI", - "doc": "https://docs.x.ai/docs/models", - "models": { - "grok-2-1212": { - "id": "grok-2-1212", - "name": "Grok 2 (1212)", - "family": "grok", - "attachment": false, - "reasoning": false, + }, + "openai/o3": { + "id": "openai/o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-12-12", - "last_updated": "2024-12-12", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { "input": 2, - "output": 10, - "cache_read": 2 + "output": 8, + "cache_read": 0.5 } }, - "grok-vision-beta": { - "id": "grok-vision-beta", - "name": "Grok Vision Beta", - "family": "grok-vision", + "openai/gpt-5.6-sol-pro": { + "id": "openai/gpt-5.6-sol-pro", + "name": "GPT-5.6 Sol Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 4096 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { "input": 5, - "output": 15, - "cache_read": 5 + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "grok-4.3": { - "id": "grok-4.3", - "name": "Grok 4.3", - "family": "grok", + "openai/o4-mini-high": { + "id": "openai/o4-mini-high", + "name": "o4 Mini High", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-05-01", - "last_updated": "2026-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-06-30", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 30000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2, - "tiers": [ - { - "input": 2.5, - "output": 5, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 5, - "cache_read": 0.4 - } + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "grok-3-mini-fast": { - "id": "grok-3-mini-fast", - "name": "Grok 3 Mini Fast", - "family": "grok", - "attachment": false, - "reasoning": true, + "openai/gpt-audio": { + "id": "openai/gpt-audio", + "name": "GPT Audio", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 4, - "reasoning": 4, - "cache_read": 0.15 + "input": 2.5, + "output": 10 } }, - "grok-3-mini-latest": { - "id": "grok-3-mini-latest", - "name": "Grok 3 Mini Latest", - "family": "grok", - "attachment": false, + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT-5.2 Pro", + "description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows", + "family": "gpt-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["image", "text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.5, - "reasoning": 0.5, - "cache_read": 0.075 + "input": 21, + "output": 168 } }, - "grok-3-fast": { - "id": "grok-3-fast", - "name": "Grok 3 Fast", - "family": "grok", + "openai/gpt-4o-mini-search-preview": { + "id": "openai/gpt-4o-mini-search-preview", + "name": "GPT-4o-mini Search Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "o-mini", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "tool_call": false, + "structured_output": true, + "temperature": false, + "knowledge": "2023-10-31", + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 1.25 + "input": 0.15, + "output": 0.6 } }, - "grok-2-vision-latest": { - "id": "grok-2-vision-latest", - "name": "Grok 2 Vision Latest", - "family": "grok", + "openai/gpt-5.6-terra-pro": { + "id": "openai/gpt-5.6-terra-pro", + "name": "GPT-5.6 Terra Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-12-12", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 4096 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 2, - "output": 10, - "cache_read": 2 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "grok-4.20-0309-reasoning": { - "id": "grok-4.20-0309-reasoning", - "name": "Grok 4.20 (Reasoning)", - "family": "grok", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-09", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - } + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", + "openai/gpt-5-chat": { + "id": "openai/gpt-5-chat", + "name": "GPT-5 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "tool_call": false, + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text", "image"], + "input": ["pdf", "image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "grok-3-mini-fast-latest": { - "id": "grok-3-mini-fast-latest", - "name": "Grok 3 Mini Fast Latest", - "family": "grok", + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2021-09-01", + "release_date": "2023-03-01", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 16385, + "output": 4096 }, "cost": { - "input": 0.6, - "output": 4, - "reasoning": 4, - "cache_read": 0.15 + "input": 0.5, + "output": 1.5 } }, - "grok-4-fast": { - "id": "grok-4-fast", - "name": "Grok 4 Fast", - "family": "grok", + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 15, + "output": 120 } }, - "grok-3-latest": { - "id": "grok-3-latest", - "name": "Grok 3 Latest", - "family": "grok", - "attachment": false, + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 2.5, + "output": 10 } }, - "grok-2": { - "id": "grok-2", - "name": "Grok 2", - "family": "grok", + "openai/gpt-4": { + "id": "openai/gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 8191, + "output": 4096 }, "cost": { - "input": 2, - "output": 10, - "cache_read": 2 + "input": 30, + "output": 60 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", - "attachment": false, + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["image", "text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "grok-4.20-0309-non-reasoning": { - "id": "grok-4.20-0309-non-reasoning", - "name": "Grok 4.20 (Non-Reasoning)", - "family": "grok", - "attachment": true, + "openai/gpt-3.5-turbo-16k": { + "id": "openai/gpt-3.5-turbo-16k", + "name": "GPT-3.5 Turbo 16k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-09", + "knowledge": "2021-09-30", + "release_date": "2023-08-28", + "last_updated": "2023-08-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 16385, + "output": 4096 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 + "input": 3, + "output": 4 + } + }, + "openai/o3-pro": { + "id": "openai/o3-pro", + "name": "o3-pro", + "description": "High-effort o3 tier for difficult technical reasoning and careful answers", + "family": "o-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-06-10", + "last_updated": "2025-06-10", + "modalities": { + "input": ["text", "pdf", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "output": 100000 + }, + "cost": { + "input": 20, + "output": 80 } }, - "grok-3-fast-latest": { - "id": "grok-3-fast-latest", - "name": "Grok 3 Fast Latest", - "family": "grok", - "attachment": false, + "openai/gpt-5.1-chat": { + "id": "openai/gpt-5.1-chat", + "name": "GPT-5.1 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["pdf", "image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 32000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 1.25 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "grok-4.20-multi-agent-0309": { - "id": "grok-4.20-multi-agent-0309", - "name": "Grok 4.20 Multi-Agent", - "family": "grok", + "openai/gpt-4o-2024-05-13": { + "id": "openai/gpt-4o-2024-05-13", + "name": "GPT-4o (2024-05-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-09", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - } - } - }, - "grok-4": { - "id": "grok-4", - "name": "Grok 4", - "family": "grok", - "attachment": false, + "input": 5, + "output": 15 + } + }, + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["pdf", "image", "text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "reasoning": 15, - "cache_read": 0.75 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "grok-2-latest": { - "id": "grok-2-latest", - "name": "Grok 2 Latest", - "family": "grok", - "attachment": false, + "openai/gpt-5.3-chat": { + "id": "openai/gpt-5.3-chat", + "name": "GPT-5.3 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-12-12", + "structured_output": true, + "temperature": false, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 10, - "cache_read": 2 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "grok-beta": { - "id": "grok-beta", - "name": "Grok Beta", - "family": "grok-beta", + "openai/gpt-3.5-turbo-0613": { + "id": "openai/gpt-3.5-turbo-0613", + "name": "GPT-3.5 Turbo (older v0613)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "knowledge": "2021-09-30", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, + "context": 4095, "output": 4096 }, "cost": { - "input": 5, - "output": 15, - "cache_read": 5 + "input": 1, + "output": 2 } }, - "grok-2-vision": { - "id": "grok-2-vision", - "name": "Grok 2 Vision", - "family": "grok", + "openai/gpt-5-image-mini": { + "id": "openai/gpt-5-image-mini", + "name": "GPT-5 Image Mini", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "release_date": "2025-10-16", + "last_updated": "2025-10-16", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["pdf", "image", "text"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 4096 + "context": 400000, + "output": 128000 }, "cost": { - "input": 2, - "output": 10, - "cache_read": 2 + "input": 2.5, + "output": 2, + "cache_read": 0.25 } }, - "grok-4-1-fast": { - "id": "grok-4-1-fast", - "name": "Grok 4.1 Fast", - "family": "grok", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Codex GPT for repository edits, code review, and practical software agents", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "grok-3-mini": { - "id": "grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", - "attachment": false, + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.5, - "reasoning": 0.5, - "cache_read": 0.075 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "grok-2-vision-1212": { - "id": "grok-2-vision-1212", - "name": "Grok 2 Vision (1212)", - "family": "grok", - "attachment": true, - "reasoning": false, + "openai/gpt-oss-120b:free": { + "id": "openai/gpt-oss-120b:free", + "name": "gpt-oss-120b (free)", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-12-12", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 4096 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2, - "output": 10, - "cache_read": 2 + "input": 0, + "output": 0 } }, - "grok-3": { - "id": "grok-3", - "name": "Grok 3", - "family": "grok", - "attachment": false, + "openai/gpt-4o-2024-08-06": { + "id": "openai/gpt-4o-2024-08-06", + "name": "GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "knowledge": "2023-09", + "release_date": "2024-08-06", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } - } - } - }, - "meganova": { - "id": "meganova", - "env": ["MEGANOVA_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.meganova.ai/v1", - "name": "Meganova", - "doc": "https://docs.meganova.ai", - "models": { - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + }, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["pdf", "image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.09, - "output": 0.6 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Qwen/Qwen3.5-Plus": { - "id": "Qwen/Qwen3.5-Plus", - "name": "Qwen3.5 Plus", - "family": "qwen", - "attachment": false, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02", - "last_updated": "2026-02", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2.4, - "reasoning": 2.4 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Qwen/Qwen2.5-VL-32B-Instruct": { - "id": "Qwen/Qwen2.5-VL-32B-Instruct", - "name": "Qwen2.5 VL 32B Instruct", - "family": "qwen", + "openai/gpt-audio-mini": { + "id": "openai/gpt-audio-mini", + "name": "GPT Audio Mini", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "o-mini", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "audio"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16384, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.6, + "output": 2.4 } }, - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "openai/gpt-5.6-luna": { + "id": "openai/gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 1, + "output": 6, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 100000 }, "cost": { - "input": 0.8, - "output": 2.56 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "GLM-4.6", - "family": "glm", - "attachment": false, + "openai/gpt-5.6-terra": { + "id": "openai/gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 1.9 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "cache_write": 3.125 } }, - "mistralai/Mistral-Small-3.2-24B-Instruct-2506": { - "id": "mistralai/Mistral-Small-3.2-24B-Instruct-2506", - "name": "Mistral Small 3.2 24B Instruct", - "family": "mistral-small", + "openai/o4-mini-deep-research": { + "id": "openai/o4-mini-deep-research", + "name": "o4-mini-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o-mini", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-06-20", - "last_updated": "2025-06-20", + "knowledge": "2024-05", + "release_date": "2024-06-26", + "last_updated": "2024-06-26", "modalities": { - "input": ["text", "image"], + "input": ["pdf", "image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "mistralai/Mistral-Nemo-Instruct-2407": { - "id": "mistralai/Mistral-Nemo-Instruct-2407", - "name": "Mistral Nemo Instruct 2407", - "family": "mistral", - "attachment": false, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["image", "text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.02, - "output": 0.04 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "XiaomiMiMo/MiMo-V2-Flash": { - "id": "XiaomiMiMo/MiMo-V2-Flash", - "name": "MiMo V2 Flash", - "family": "mimo", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.036, + "output": 0.18 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", - "attachment": false, + "openai/gpt-4o-2024-11-20": { + "id": "openai/gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2023-09", + "release_date": "2024-11-20", + "last_updated": "2024-11-20", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "openai/o1": { + "id": "openai/o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-25", - "last_updated": "2025-08-25", + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.27, - "output": 1 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "deepseek-ai/DeepSeek-V3-0324": { - "id": "deepseek-ai/DeepSeek-V3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "openai/o1-pro": { + "id": "openai/o1-pro", + "name": "o1-pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": false, + "knowledge": "2023-09", + "release_date": "2025-03-19", + "last_updated": "2025-03-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.25, - "output": 0.88 + "input": 150, + "output": 600 } }, - "deepseek-ai/DeepSeek-V3.2-Exp": { - "id": "deepseek-ai/DeepSeek-V3.2-Exp", - "name": "DeepSeek V3.2 Exp", - "family": "deepseek", - "attachment": false, + "openai/gpt-chat-latest": { + "id": "openai/gpt-chat-latest", + "name": "GPT Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-10", - "last_updated": "2025-10-10", + "temperature": false, + "release_date": "2026-05-05", + "last_updated": "2026-05-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 0.4 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "deepseek-ai/DeepSeek-R1-0528": { - "id": "deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek-thinking", - "attachment": false, + "openai/gpt-5-image": { + "id": "openai/gpt-5-image", + "name": "GPT-5 Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": false, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "knowledge": "2024-10-01", + "release_date": "2025-10-14", + "last_updated": "2025-10-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["image", "text", "pdf"], + "output": ["image", "text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 64000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 2.15 + "input": 10, + "output": 10, + "cache_read": 1.25 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, - "reasoning": false, + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.26, - "output": 0.38 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["pdf", "image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.6 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, - "reasoning": true, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.45, - "output": 2.8 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "openai/o3-deep-research": { + "id": "openai/o3-deep-research", + "name": "o3-deep-research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-05", + "release_date": "2024-06-26", + "last_updated": "2024-06-26", "modalities": { - "input": ["text"], + "input": ["image", "text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 10, + "output": 40, + "cache_read": 2.5 } }, - "MiniMaxAI/MiniMax-M2.1": { - "id": "MiniMaxAI/MiniMax-M2.1", - "name": "MiniMax M2.1", - "family": "minimax", + "openai/gpt-4-turbo-preview": { + "id": "openai/gpt-4-turbo-preview", + "name": "GPT-4 Turbo Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2023-12-31", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 131072 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.28, - "output": 1.2 + "input": 10, + "output": 30 } - } - } - }, - "google-vertex-anthropic": { - "id": "google-vertex-anthropic", - "env": ["GOOGLE_VERTEX_PROJECT", "GOOGLE_VERTEX_LOCATION", "GOOGLE_APPLICATION_CREDENTIALS"], - "npm": "@ai-sdk/google-vertex/anthropic", - "name": "Vertex (Anthropic)", - "doc": "https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude", - "models": { - "claude-haiku-4-5@20251001": { - "id": "claude-haiku-4-5@20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "claude-sonnet-4-6@default": { - "id": "claude-sonnet-4-6@default", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - }, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "claude-3-5-haiku@20241022": { - "id": "claude-3-5-haiku@20241022", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 10, + "output": 30 } }, - "claude-3-5-sonnet@20241022": { - "id": "claude-3-5-sonnet@20241022", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01 } }, - "claude-opus-4-1@20250805": { - "id": "claude-opus-4-1@20250805", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1050000, + "input": 922000, + "output": 128000 }, - "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "cost": { + "input": 30, + "output": 180 } }, - "claude-sonnet-4@20250514": { - "id": "claude-sonnet-4@20250514", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "openai/o3-mini-high": { + "id": "openai/o3-mini-high", + "name": "o3 Mini High", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": true, + "temperature": false, + "knowledge": "2023-10-31", + "release_date": "2025-02-12", + "last_updated": "2025-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 64000 + "output": 100000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "claude-3-7-sonnet@20250219": { - "id": "claude-3-7-sonnet@20250219", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", + "openai/gpt-5.4-image-2": { + "id": "openai/gpt-5.4-image-2", + "name": "GPT-5.4 Image 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": false, + "structured_output": true, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["image", "text", "pdf"], + "output": ["image", "text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 272000, + "output": 128000 }, "cost": { - "input": 3, + "input": 8, "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "cache_read": 2 } }, - "claude-opus-4@20250514": { - "id": "claude-opus-4@20250514", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "openai/gpt-4o-search-preview": { + "id": "openai/gpt-4o-search-preview", + "name": "GPT-4o Search Preview", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": true, + "temperature": false, + "knowledge": "2023-10-31", + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 2.5, + "output": 10 } }, - "claude-opus-4-5@20251101": { - "id": "claude-opus-4-5@20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 30, + "output": 180 } }, - "claude-sonnet-4-5@20250929": { - "id": "claude-sonnet-4-5@20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "claude-opus-4-6@default": { - "id": "claude-opus-4-6@default", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - }, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.029, + "output": 0.14 } }, - "claude-opus-4-7@default": { - "id": "claude-opus-4-7@default", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "openai/gpt-5.6-sol": { + "id": "openai/gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "structured_output": true, "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { "input": 5, - "output": 25, + "output": 30, "cache_read": 0.5, - "cache_write": 6.25, - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - }, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "cache_write": 6.25 } - } - } - }, - "evroc": { - "id": "evroc", - "env": ["EVROC_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://models.think.evroc.com/v1", - "name": "evroc", - "doc": "https://docs.evroc.com/products/think/overview.html", - "models": { - "Qwen/Qwen3-VL-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", - "name": "Qwen3 VL 30B", - "family": "qwen", - "attachment": false, - "reasoning": false, + }, + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 100000, - "output": 100000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.24, - "output": 0.94 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "Qwen/Qwen3-Embedding-8B": { - "id": "Qwen/Qwen3-Embedding-8B", - "name": "Qwen3 Embedding 8B", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 40960, - "output": 40960 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0.12 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Qwen/Qwen3-30B-A3B-Instruct-2507-FP8": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507-FP8", - "name": "Qwen3 30B 2507", - "family": "qwen", + "openai/gpt-oss-20b:free": { + "id": "openai/gpt-oss-20b:free", + "name": "gpt-oss-20b (free)", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "structured_output": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 64000, - "output": 64000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.35, - "output": 1.42 + "input": 0, + "output": 0 } }, - "mistralai/devstral-small-2-24b-instruct-2512": { - "id": "mistralai/devstral-small-2-24b-instruct-2512", - "name": "Devstral Small 2 24B Instruct 2512", - "family": "devstral", - "attachment": false, - "reasoning": false, + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["image", "text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0.47 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "mistralai/Voxtral-Small-24B-2507": { - "id": "mistralai/Voxtral-Small-24B-2507", - "name": "Voxtral Small 24B", - "family": "voxtral", - "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["audio", "text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 32000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.00236, - "output": 0.00236, - "output_audio": 2.36 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "mistralai/Magistral-Small-2509": { - "id": "mistralai/Magistral-Small-2509", - "name": "Magistral Small 1.2 24B", - "family": "magistral-small", + "thedrummer/cydonia-24b-v4.1": { + "id": "thedrummer/cydonia-24b-v4.1", + "name": "Cydonia 24B V4.1", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2025-06-01", - "last_updated": "2025-06-01", + "structured_output": true, + "temperature": true, + "knowledge": "2024-04-30", + "release_date": "2025-09-27", + "last_updated": "2025-09-27", "modalities": { "input": ["text"], "output": ["text"] @@ -79421,113 +128686,131 @@ "output": 131072 }, "cost": { - "input": 0.59, - "output": 2.36 + "input": 0.3, + "output": 0.5, + "cache_read": 0.15 } }, - "microsoft/Phi-4-multimodal-instruct": { - "id": "microsoft/Phi-4-multimodal-instruct", - "name": "Phi-4 15B", - "family": "phi", + "thedrummer/skyfall-36b-v2": { + "id": "thedrummer/skyfall-36b-v2", + "name": "Skyfall 36B V2", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "structured_output": true, + "temperature": true, + "knowledge": "2024-06-30", + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 32000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.24, - "output": 0.47 + "input": 0.55, + "output": 0.8, + "cache_read": 0.25 } }, - "KBLab/kb-whisper-large": { - "id": "KBLab/kb-whisper-large", - "name": "KB Whisper", - "family": "whisper", + "thedrummer/unslopnemo-12b": { + "id": "thedrummer/unslopnemo-12b", + "name": "UnslopNemo 12B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": false, - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-04-30", + "release_date": "2024-11-08", + "last_updated": "2024-11-08", "modalities": { - "input": ["audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 448, - "output": 448 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.00236, - "output": 0.00236, - "output_audio": 2.36 + "input": 0.4, + "output": 0.4 } }, - "nvidia/Llama-3.3-70B-Instruct-FP8": { - "id": "nvidia/Llama-3.3-70B-Instruct-FP8", - "name": "Llama 3.3 70B", - "family": "llama", + "thedrummer/rocinante-12b": { + "id": "thedrummer/rocinante-12b", + "name": "Rocinante 12B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "structured_output": true, + "temperature": true, + "knowledge": "2024-04-30", + "release_date": "2024-09-30", + "last_updated": "2024-09-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 65536, + "output": 65536 }, "cost": { - "input": 1.18, - "output": 1.18 + "input": 0.25, + "output": 0.5 } }, - "openai/whisper-large-v3": { - "id": "openai/whisper-large-v3", - "name": "Whisper 3 Large", - "family": "whisper", - "attachment": false, + "bytedance/ui-tars-1.5-7b": { + "id": "bytedance/ui-tars-1.5-7b", + "name": "UI-TARS 7B ", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, "reasoning": false, "tool_call": false, - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "structured_output": true, + "temperature": true, + "knowledge": "2025-01-31", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { - "input": ["audio"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 448, - "output": 4096 + "context": 128000, + "output": 2048 }, "cost": { - "input": 0.00236, - "output": 0.00236, - "output_audio": 2.36 + "input": 0.1, + "output": 0.2, + "cache_read": 0.1 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "rekaai/reka-flash-3": { + "id": "rekaai/reka-flash-3", + "name": "Reka Flash 3", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "reka", "attachment": false, "reasoning": true, - "tool_call": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01-31", + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text"], "output": ["text"] @@ -79538,1528 +128821,1974 @@ "output": 65536 }, "cost": { - "input": 0.24, - "output": 0.94 + "input": 0.1, + "output": 0.2 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, - "reasoning": true, + "rekaai/reka-edge": { + "id": "rekaai/reka-edge", + "name": "Reka Edge", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "reka", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "structured_output": true, + "temperature": true, + "release_date": "2026-03-20", + "last_updated": "2026-03-20", "modalities": { - "input": ["text", "image", "video"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 16384, + "output": 16384 }, "cost": { - "input": 1.47, - "output": 5.9 + "input": 0.1, + "output": 0.1 } }, - "intfloat/multilingual-e5-large-instruct": { - "id": "intfloat/multilingual-e5-large-instruct", - "name": "E5 Multi-Lingual Large Embeddings 0.6B", - "family": "text-embedding", - "attachment": false, + "mistralai/mistral-large-2407": { + "id": "mistralai/mistral-large-2407", + "name": "Mistral Large 2407", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, "reasoning": false, - "tool_call": false, - "release_date": "2024-06-01", - "last_updated": "2024-06-01", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-03-31", + "release_date": "2024-11-19", + "last_updated": "2024-11-19", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512, - "output": 512 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.12, - "output": 0.12 + "input": 2, + "output": 6, + "cache_read": 0.2 } - } - } - }, - "synthetic": { - "id": "synthetic", - "env": ["SYNTHETIC_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.synthetic.new/openai/v1", - "name": "Synthetic", - "doc": "https://synthetic.new/pricing", - "models": { - "hf:meta-llama/Llama-3.1-405B-Instruct": { - "id": "hf:meta-llama/Llama-3.1-405B-Instruct", - "name": "Llama-3.1-405B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": true, + }, + "mistralai/mistral-small-3.2-24b-instruct": { + "id": "mistralai/mistral-small-3.2-24b-instruct", + "name": "Mistral Small 3.2 24B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2023-10-31", + "release_date": "2025-06-20", + "last_updated": "2025-06-20", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 128000, - "output": 32768 + "output": 16384 }, "cost": { - "input": 3, - "output": 3 + "input": 0.075, + "output": 0.2 } }, - "hf:meta-llama/Llama-4-Scout-17B-16E-Instruct": { - "id": "hf:meta-llama/Llama-4-Scout-17B-16E-Instruct", - "name": "Llama-4-Scout-17B-16E-Instruct", - "family": "llama", - "attachment": true, + "mistralai/mistral-nemo": { + "id": "mistralai/mistral-nemo", + "name": "Mistral Nemo", + "description": "Efficient Mistral-NVIDIA open model for multilingual chat and local deployment", + "family": "mistral-nemo", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 328000, - "output": 4096 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.02, + "output": 0.03 } }, - "hf:meta-llama/Llama-3.3-70B-Instruct": { - "id": "hf:meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": false, + "mistralai/mistral-medium-3-5": { + "id": "mistralai/mistral-medium-3-5", + "name": "Mistral Medium 3.5", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.9, - "output": 0.9 + "input": 1.5, + "output": 7.5 } }, - "hf:meta-llama/Llama-3.1-8B-Instruct": { - "id": "hf:meta-llama/Llama-3.1-8B-Instruct", - "name": "Llama-3.1-8B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": true, + "mistralai/ministral-8b-2512": { + "id": "mistralai/ministral-8b-2512", + "name": "Ministral 3 8B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.15, + "output": 0.15, + "cache_read": 0.015 } }, - "hf:meta-llama/Llama-3.1-70B-Instruct": { - "id": "hf:meta-llama/Llama-3.1-70B-Instruct", - "name": "Llama-3.1-70B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": true, - "tool_call": true, + "mistralai/mistral-small-3.1-24b-instruct": { + "id": "mistralai/mistral-small-3.1-24b-instruct", + "name": "Mistral Small 3.1 24B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2023-10-31", + "release_date": "2025-03-17", + "last_updated": "2025-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { "context": 128000, - "output": 32768 + "output": 128000 }, "cost": { - "input": 0.9, - "output": 0.9 + "input": 0.351, + "output": 0.555 } }, - "hf:meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": { - "id": "hf:meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8", - "name": "Llama-4-Maverick-17B-128E-Instruct-FP8", - "family": "llama", + "mistralai/mistral-saba": { + "id": "mistralai/mistral-saba", + "name": "Saba", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "knowledge": "2024-09-30", + "release_date": "2025-02-17", + "last_updated": "2025-02-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 524000, - "output": 4096 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.22, - "output": 0.88 + "input": 0.2, + "output": 0.6, + "cache_read": 0.02 } }, - "hf:MiniMaxAI/MiniMax-M2": { - "id": "hf:MiniMaxAI/MiniMax-M2", - "name": "MiniMax-M2", - "family": "minimax", - "attachment": false, - "reasoning": true, + "mistralai/mistral-large": { + "id": "mistralai/mistral-large", + "name": "Mistral Large", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2024-11-30", + "release_date": "2024-02-26", + "last_updated": "2024-02-26", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 131000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 2, + "output": 6, + "cache_read": 0.2 } }, - "hf:MiniMaxAI/MiniMax-M2.5": { - "id": "hf:MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, + "mistralai/mistral-medium-3.1": { + "id": "mistralai/mistral-medium-3.1", + "name": "Mistral Medium 3.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-02-07", - "last_updated": "2026-02-07", + "knowledge": "2025-06-30", + "release_date": "2025-08-13", + "last_updated": "2025-08-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 191488, - "output": 65536 + "context": 131072, + "output": 262144 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.6 + "input": 0.4, + "output": 2, + "cache_read": 0.04 } }, - "hf:MiniMaxAI/MiniMax-M2.1": { - "id": "hf:MiniMaxAI/MiniMax-M2.1", - "name": "MiniMax-M2.1", - "family": "minimax", + "mistralai/mistral-small-24b-instruct-2501": { + "id": "mistralai/mistral-small-24b-instruct-2501", + "name": "Mistral Small 3", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2023-10-31", + "release_date": "2025-01-30", + "last_updated": "2025-01-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0.05, + "output": 0.08 } }, - "hf:Qwen/Qwen3.5-397B-A17B": { - "id": "hf:Qwen/Qwen3.5-397B-A17B", - "name": "Qwen3.5-97B-A17B", - "family": "qwen", - "attachment": false, - "reasoning": true, + "mistralai/ministral-3b-2512": { + "id": "mistralai/ministral-3b-2512", + "name": "Ministral 3 3B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 131072 }, - "status": "beta", "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.6 + "input": 0.1, + "output": 0.1, + "cache_read": 0.01 } }, - "hf:Qwen/Qwen2.5-Coder-32B-Instruct": { - "id": "hf:Qwen/Qwen2.5-Coder-32B-Instruct", - "name": "Qwen2.5-Coder-32B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, + "mistralai/mistral-small-2603": { + "id": "mistralai/mistral-small-2603", + "name": "Mistral Small 4", + "description": "Fast Mistral production model for chat, extraction, and cost-sensitive agents", + "family": "mistral-small", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-11-11", - "last_updated": "2024-11-11", + "knowledge": "2025-06", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.8, - "output": 0.8 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "hf:Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "hf:Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen 3 235B Instruct", - "family": "qwen", - "attachment": false, + "mistralai/ministral-14b-2512": { + "id": "mistralai/ministral-14b-2512", + "name": "Ministral 3 14B 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-28", - "last_updated": "2025-07-21", + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { "input": 0.2, - "output": 0.6 + "output": 0.2, + "cache_read": 0.02 } }, - "hf:Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "hf:Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3 235B A22B Thinking 2507", - "family": "qwen", - "attachment": false, - "reasoning": true, + "mistralai/devstral-2512": { + "id": "mistralai/devstral-2512", + "name": "Devstral 2", + "description": "Mistral's coding-agent model for repository work, terminal tasks, and software fixes", + "family": "devstral", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "knowledge": "2025-12", + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 262144, + "output": 262144 }, + "status": "deprecated", "cost": { - "input": 0.65, - "output": 3 + "input": 0.4, + "output": 2, + "cache_read": 0.04 } }, - "hf:Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "hf:Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen 3 Coder 480B", - "family": "qwen", - "attachment": false, + "mistralai/mixtral-8x22b-instruct": { + "id": "mistralai/mixtral-8x22b-instruct", + "name": "Mixtral 8x22B Instruct", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2024-01-31", + "release_date": "2024-04-17", + "last_updated": "2024-04-17", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 65536, + "output": 65536 }, "cost": { "input": 2, - "output": 2 + "output": 6, + "cache_read": 0.2 } }, - "hf:deepseek-ai/DeepSeek-V3.1": { - "id": "hf:deepseek-ai/DeepSeek-V3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": false, - "reasoning": true, + "mistralai/mistral-medium-3": { + "id": "mistralai/mistral-medium-3", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "knowledge": "2025-03-31", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.56, - "output": 1.68 + "input": 0.4, + "output": 2, + "cache_read": 0.04 } }, - "hf:deepseek-ai/DeepSeek-V3-0324": { - "id": "hf:deepseek-ai/DeepSeek-V3-0324", - "name": "DeepSeek V3 (0324)", - "family": "deepseek", - "attachment": false, + "mistralai/voxtral-small-24b-2507": { + "id": "mistralai/voxtral-small-24b-2507", + "name": "Voxtral Small 24B 2507", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "release_date": "2025-10-30", + "last_updated": "2025-10-30", "modalities": { - "input": ["text"], + "input": ["text", "audio", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 32000, + "output": 32000 }, "cost": { - "input": 1.2, - "output": 1.2 + "input": 0.1, + "output": 0.3, + "cache_read": 0.01 } }, - "hf:deepseek-ai/DeepSeek-V3": { - "id": "hf:deepseek-ai/DeepSeek-V3", - "name": "DeepSeek V3", - "family": "deepseek", - "attachment": false, - "reasoning": true, + "mistralai/mistral-large-2512": { + "id": "mistralai/mistral-large-2512", + "name": "Mistral Large 3", + "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning", + "family": "mistral-large", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-29", + "knowledge": "2024-11", + "release_date": "2024-11-01", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.25, - "output": 1.25 + "input": 0.5, + "output": 1.5, + "cache_read": 0.05 } }, - "hf:deepseek-ai/DeepSeek-R1": { - "id": "hf:deepseek-ai/DeepSeek-R1", - "name": "DeepSeek R1", - "family": "deepseek-thinking", - "attachment": false, - "reasoning": true, + "mistralai/codestral-2508": { + "id": "mistralai/codestral-2508", + "name": "Codestral 2508", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "codestral", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "knowledge": "2025-03-31", + "release_date": "2025-08-01", + "last_updated": "2025-08-01", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0.3, + "output": 0.9, + "cache_read": 0.03 } }, - "hf:deepseek-ai/DeepSeek-R1-0528": { - "id": "hf:deepseek-ai/DeepSeek-R1-0528", - "name": "DeepSeek R1 (0528)", - "family": "deepseek-thinking", + "morph/morph-v3-fast": { + "id": "morph/morph-v3-fast", + "name": "Morph V3 Fast", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "morph", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "release_date": "2025-07-07", + "last_updated": "2025-07-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 81920, + "output": 38000 }, "cost": { - "input": 3, - "output": 8 + "input": 0.8, + "output": 1.2 } }, - "hf:deepseek-ai/DeepSeek-V3.2": { - "id": "hf:deepseek-ai/DeepSeek-V3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "morph/morph-v3-large": { + "id": "morph/morph-v3-large", + "name": "Morph V3 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "morph", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": true, "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2025-07-07", + "last_updated": "2025-07-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 162816, - "input": 162816, - "output": 8000 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.27, - "output": 0.4, - "cache_read": 0.27, - "cache_write": 0 + "input": 0.9, + "output": 1.9 } }, - "hf:deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "hf:deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", - "attachment": false, + "bytedance-seed/seed-1.6-flash": { + "id": "bytedance-seed/seed-1.6-flash", + "name": "Seed 1.6 Flash", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-09-22", - "last_updated": "2025-09-25", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 1.2, - "output": 1.2 + "input": 0.075, + "output": 0.3 } }, - "hf:openai/gpt-oss-120b": { - "id": "hf:openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + "bytedance-seed/seed-1.6": { + "id": "bytedance-seed/seed-1.6", + "name": "Seed 1.6", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text"], + "input": ["image", "text", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 262144, "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.25, + "output": 2 } }, - "hf:moonshotai/Kimi-K2-Thinking": { - "id": "hf:moonshotai/Kimi-K2-Thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "bytedance-seed/seed-2.0-mini": { + "id": "bytedance-seed/seed-2.0-mini", + "name": "Seed-2.0-Mini", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-07", - "last_updated": "2025-11-07", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 262144 + "output": 131072 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0.1, + "output": 0.4 } }, - "hf:moonshotai/Kimi-K2-Instruct-0905": { - "id": "hf:moonshotai/Kimi-K2-Instruct-0905", - "name": "Kimi K2 0905", - "family": "kimi", - "attachment": false, - "reasoning": false, + "bytedance-seed/seed-2.0-lite": { + "id": "bytedance-seed/seed-2.0-lite", + "name": "Seed-2.0-Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "release_date": "2026-03-10", + "last_updated": "2026-03-10", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 32768 + "output": 131072 }, "cost": { - "input": 1.2, - "output": 1.2 + "input": 0.25, + "output": 2 } }, - "hf:moonshotai/Kimi-K2.5": { - "id": "hf:moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, + "sakana/fugu-ultra": { + "id": "sakana/fugu-ultra", + "name": "Fugu Ultra", + "description": "Quality-first multi-agent model for hard research, analysis, and competitions", + "family": "fugu", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["max", "xhigh", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "structured_output": true, + "temperature": false, + "release_date": "2026-06-24", + "last_updated": "2026-06-24", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4": { - "id": "hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4", - "name": "Nemotron 3 Super 120B", - "family": "nemotron", + "anthracite-org/magnum-v4-72b": { + "id": "anthracite-org/magnum-v4-72b", + "name": "Magnum v4 72B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning": false, + "tool_call": false, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-03-11", - "last_updated": "2026-04-03", + "knowledge": "2024-06-30", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 32768, + "output": 2048 }, "cost": { - "input": 0.3, - "output": 1, - "cache_read": 0.3 + "input": 3, + "output": 5 } }, - "hf:nvidia/Kimi-K2.5-NVFP4": { - "id": "hf:nvidia/Kimi-K2.5-NVFP4", - "name": "Kimi K2.5 (NVFP4)", - "family": "kimi", + "nvidia/nemotron-3-nano-30b-a3b:free": { + "id": "nvidia/nemotron-3-nano-30b-a3b:free", + "name": "Nemotron 3 Nano 30B A3B (free)", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0, + "output": 0 } }, - "hf:zai-org/GLM-4.7-Flash": { - "id": "hf:zai-org/GLM-4.7-Flash", - "name": "GLM-4.7-Flash", - "family": "glm", + "nvidia/nemotron-nano-9b-v2:free": { + "id": "nvidia/nemotron-nano-9b-v2:free", + "name": "Nemotron Nano 9B V2 (free)", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-01-18", - "last_updated": "2026-01-18", + "release_date": "2025-08-18", + "last_updated": "2025-08-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 65536 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.06, - "output": 0.4, - "cache_read": 0.06 + "input": 0, + "output": 0 } }, - "hf:zai-org/GLM-4.7": { - "id": "hf:zai-org/GLM-4.7", - "name": "GLM 4.7", - "family": "glm", - "attachment": false, + "nvidia/nemotron-nano-12b-v2-vl:free": { + "id": "nvidia/nemotron-nano-12b-v2-vl:free", + "name": "Nemotron Nano 12B 2 VL (free)", + "description": "Nemotron multimodal model for visual reasoning and agentic AI workflows", + "family": "nemotron", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0, + "output": 0 } }, - "hf:zai-org/GLM-5.1": { - "id": "hf:zai-org/GLM-5.1", - "name": "GLM 5.1", - "family": "glm", - "attachment": false, + "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": { + "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free", + "name": "Nemotron 3 Nano Omni (free)", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-04-12", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, + "context": 256000, "output": 65536 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 1 + "input": 0, + "output": 0 } }, - "hf:zai-org/GLM-5": { - "id": "hf:zai-org/GLM-5", - "name": "GLM-5", - "family": "glm", + "nvidia/nemotron-3-ultra-550b-a55b:free": { + "id": "nvidia/nemotron-3-ultra-550b-a55b:free", + "name": "Nemotron 3 Ultra (free)", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-04-08", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, + "context": 1000000, "output": 65536 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 1 + "input": 0, + "output": 0 } }, - "hf:zai-org/GLM-4.6": { - "id": "hf:zai-org/GLM-4.6", - "name": "GLM 4.6", - "family": "glm", + "nvidia/nemotron-3-ultra-550b-a55b": { + "id": "nvidia/nemotron-3-ultra-550b-a55b", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high"] + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.19 + "input": 0.5, + "output": 2.2, + "cache_read": 0.1 } }, - "hf:moonshotai/Kimi-K2.6": { - "id": "hf:moonshotai/Kimi-K2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "nvidia/nemotron-3.5-content-safety:free": { + "id": "nvidia/nemotron-3.5-content-safety:free", + "name": "Nemotron 3.5 Content Safety (free)", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "nemotron", "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.95 + "input": 0, + "output": 0 } - } - } - }, - "nvidia": { - "id": "nvidia", - "env": ["NVIDIA_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://integrate.api.nvidia.com/v1", - "name": "Nvidia", - "doc": "https://docs.api.nvidia.com/nim/", - "models": { - "upstage/solar-10_7b-instruct": { - "id": "upstage/solar-10_7b-instruct", - "name": "solar-10.7b-instruct", + }, + "nvidia/nemotron-3-nano-30b-a3b": { + "id": "nvidia/nemotron-3-nano-30b-a3b", + "name": "Nemotron 3 Nano 30B A3B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-06-05", - "last_updated": "2025-04-10", + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 228000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.05, + "output": 0.2 } }, - "black-forest-labs/flux_2-klein-4b": { - "id": "black-forest-labs/flux_2-klein-4b", - "name": "FLUX.2 Klein 4B", - "family": "flux", + "nvidia/llama-3.3-nemotron-super-49b-v1.5": { + "id": "nvidia/llama-3.3-nemotron-super-49b-v1.5", + "name": "Llama 3.3 Nemotron Super 49B v1.5", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-01-14", - "last_updated": "2026-01-31", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["image", "text"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 40960 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.4, + "output": 0.4 } }, - "black-forest-labs/flux.1-dev": { - "id": "black-forest-labs/flux.1-dev", - "name": "FLUX.1-dev", - "family": "flux", + "nvidia/nemotron-3-super-120b-a12b:free": { + "id": "nvidia/nemotron-3-super-120b-a12b:free", + "name": "Nemotron 3 Super (free)", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-01", - "last_updated": "2025-09-05", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 4096, - "output": 0 + "context": 1000000, + "output": 262144 }, "cost": { "input": 0, "output": 0 } }, - "black-forest-labs/flux_1-schnell": { - "id": "black-forest-labs/flux_1-schnell", - "name": "FLUX.1-schnell", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": false, - "knowledge": "2024-07", - "release_date": "2024-08-01", - "last_updated": "2026-02-04", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium"] + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": true, "limit": { - "context": 77, - "input": 77, - "output": 0 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.08, + "output": 0.45 } }, - "black-forest-labs/flux_1-kontext-dev": { - "id": "black-forest-labs/flux_1-kontext-dev", - "name": "FLUX.1-Kontext-dev", - "attachment": true, + "cognitivecomputations/dolphin-mistral-24b-venice-edition": { + "id": "cognitivecomputations/dolphin-mistral-24b-venice-edition", + "name": "Uncensored", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "structured_output": false, + "temperature": true, + "knowledge": "2024-04-30", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 40960, - "output": 40960 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 0.9 } }, - "stepfun-ai/step-3.5-flash": { - "id": "stepfun-ai/step-3.5-flash", - "name": "Step 3.5 Flash", + "cognitivecomputations/dolphin-mistral-24b-venice-edition:free": { + "id": "cognitivecomputations/dolphin-mistral-24b-venice-edition:free", + "name": "Uncensored (free)", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2026-02-02", - "last_updated": "2026-02-02", + "knowledge": "2024-04-30", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 32768, + "output": 32768 }, "cost": { "input": 0, "output": 0 } }, - "mistralai/mistral-large-3-675b-instruct-2512": { - "id": "mistralai/mistral-large-3-675b-instruct-2512", - "name": "Mistral Large 3 675B Instruct 2512", - "family": "mistral-large", + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.105, + "output": 0.28, + "cache_read": 0.028 } }, - "mistralai/devstral-2-123b-instruct-2512": { - "id": "mistralai/devstral-2-123b-instruct-2512", - "name": "Devstral-2-123B-Instruct-2512", - "family": "devstral", - "attachment": true, + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-08", - "last_updated": "2025-12-09", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } }, - "mistralai/mistral-nemotron": { - "id": "mistralai/mistral-nemotron", - "name": "mistral-nemotron", + "inception/mercury-2": { + "id": "inception/mercury-2", + "name": "Mercury 2", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "mercury", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-06-11", - "last_updated": "2025-06-12", + "release_date": "2026-03-04", + "last_updated": "2026-03-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "output": 50000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 0.75, + "cache_read": 0.025 } }, - "mistralai/mixtral-8x22b-instruct": { - "id": "mistralai/mixtral-8x22b-instruct", - "name": "Mistral: Mixtral 8x22B Instruct", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4.5": { + "id": "anthropic/claude-sonnet-4.5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-04-17", - "last_updated": "2024-04-17", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 13108 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "mistralai/mistral-medium-3-instruct": { - "id": "mistralai/mistral-medium-3-instruct", - "name": "Mistral Medium 3", - "family": "mistral-medium", + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "temperature": true, + "knowledge": "2025-01-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["image", "text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 131072, - "output": 32768 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "mistralai/mixtral-8x7b-instruct": { - "id": "mistralai/mixtral-8x7b-instruct", - "name": "Mistral: Mixtral 8x7B Instruct", - "attachment": false, - "reasoning": false, + "anthropic/claude-haiku-4.5": { + "id": "anthropic/claude-haiku-4.5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2023-12-10", - "last_updated": "2026-03-15", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 16384 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "mistralai/magistral-small-2506": { - "id": "mistralai/magistral-small-2506", - "name": "Magistral Small 2506", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "anthropic/claude-opus-4.7-fast": { + "id": "anthropic/claude-opus-4.7-fast", + "name": "Claude Opus 4.7 (Fast)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "input": 32768, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 } }, - "mistralai/mistral-7b-instruct-v03": { - "id": "mistralai/mistral-7b-instruct-v03", - "name": "Mistral-7B-Instruct-v0.3", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-5": { + "id": "anthropic/claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-01", - "last_updated": "2025-04-01", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "mistralai/mistral-small-4-119b-2603": { - "id": "mistralai/mistral-small-4-119b-2603", - "name": "mistral-small-4-119b-2603", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "sarvamai/sarvam-m": { - "id": "sarvamai/sarvam-m", - "name": "sarvam-m", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "microsoft/phi-4-mini-instruct": { - "id": "microsoft/phi-4-mini-instruct", - "name": "Phi-4-Mini", - "family": "phi", + "anthropic/claude-fable-5": { + "id": "anthropic/claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2025-09-05", + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "microsoft/phi-4-multimodal-instruct": { - "id": "microsoft/phi-4-multimodal-instruct", - "name": "Phi 4 Multimodal", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 16384 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": { - "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning", - "name": "Nemotron 3 Nano Omni", - "family": "nemotron", + "anthropic/claude-opus-4.5": { + "id": "anthropic/claude-opus-4.5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-28", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "nvidia/usdvalidate": { - "id": "nvidia/usdvalidate", - "name": "usdvalidate", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-07-24", - "last_updated": "2025-01-08", + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "nvidia/synthetic-video-detector": { - "id": "nvidia/synthetic-video-detector", - "name": "synthetic-video-detector", + "anthropic/claude-3-haiku": { + "id": "anthropic/claude-3-haiku", + "name": "Claude 3 Haiku", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude", "attachment": true, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "knowledge": "2023-08-31", + "release_date": "2024-03-13", + "last_updated": "2024-03-13", "modalities": { - "input": ["video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, + "context": 200000, "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 1.25, + "cache_read": 0.03, + "cache_write": 0.3 } }, - "nvidia/cosmos-transfer1-7b": { - "id": "nvidia/cosmos-transfer1-7b", - "name": "cosmos-transfer1-7b", + "anthropic/claude-opus-4": { + "id": "anthropic/claude-opus-4", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-06-13", - "last_updated": "2025-06-30", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2025-01-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image", "video"], - "output": ["video"] + "input": ["image", "text", "pdf"], + "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 4096 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "nvidia/rerank-qa-mistral-4b": { - "id": "nvidia/rerank-qa-mistral-4b", - "name": "rerank-qa-mistral-4b", - "attachment": false, - "reasoning": false, - "tool_call": false, + "anthropic/claude-opus-4.8-fast": { + "id": "anthropic/claude-opus-4.8-fast", + "name": "Claude Opus 4.8 (Fast)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2024-03-17", - "last_updated": "2025-01-17", + "release_date": "2026-05-27", + "last_updated": "2026-05-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "nvidia/nv-embedcode-7b-v1": { - "id": "nvidia/nv-embedcode-7b-v1", - "name": "nv-embedcode-7b-v1", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-03-17", - "last_updated": "2025-05-29", + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 2048 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "Nemotron 3 Super", - "family": "nemotron", + "tencent/hy3:free": { + "id": "tencent/hy3:free", + "name": "Hy3 (free)", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hy3", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "release_date": "2026-07-06", + "last_updated": "2026-07-06", "modalities": { "input": ["text"], "output": ["text"] @@ -81070,90 +130799,117 @@ "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0, + "output": 0 } }, - "nvidia/riva-translate-4b-instruct-v1_1": { - "id": "nvidia/riva-translate-4b-instruct-v1_1", - "name": "riva-translate-4b-instruct-v1_1", + "tencent/hy3": { + "id": "tencent/hy3", + "name": "Hy3", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hy3", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-12-12", - "last_updated": "2025-12-12", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-07-06", + "last_updated": "2026-07-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.14, + "output": 0.58, + "cache_read": 0.035 } }, - "nvidia/nemotron-voicechat": { - "id": "nvidia/nemotron-voicechat", - "name": "nemotron-voicechat", - "attachment": true, - "reasoning": false, - "tool_call": true, + "tencent/hunyuan-a13b-instruct": { + "id": "tencent/hunyuan-a13b-instruct", + "name": "Hunyuan A13B Instruct", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "hunyuan", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-03-31", + "release_date": "2025-07-08", + "last_updated": "2025-07-08", "modalities": { - "input": ["text", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.14, + "output": 0.57 } }, - "nvidia/llama-3_3-nemotron-super-49b-v1_5": { - "id": "nvidia/llama-3_3-nemotron-super-49b-v1_5", - "name": "Llama 3.3 Nemotron Super 49B v1.5", - "family": "nemotron", + "tencent/hy3-preview": { + "id": "tencent/hy3-preview", + "name": "Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", + "family": "Hy", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "high"] + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.063, + "output": 0.21, + "cache_read": 0.021 } }, - "nvidia/usdcode": { - "id": "nvidia/usdcode", - "name": "usdcode", + "deepcogito/cogito-v2.1-671b": { + "id": "deepcogito/cogito-v2.1-671b", + "name": "Cogito v2.1 671B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "cogito", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2026-01-01", - "last_updated": "2026-01-01", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text"], "output": ["text"] @@ -81161,45 +130917,53 @@ "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 1.25 } }, - "nvidia/llama-3_2-nemoretriever-300m-embed-v1": { - "id": "nvidia/llama-3_2-nemoretriever-300m-embed-v1", - "name": "llama-3_2-nemoretriever-300m-embed-v1", + "cohere/command-a": { + "id": "cohere/command-a", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-07-24", - "last_updated": "2025-07-24", + "structured_output": true, + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 2048 + "context": 256000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 10 } }, - "nvidia/llama-3_1-nemotron-safety-guard-8b-v3": { - "id": "nvidia/llama-3_1-nemotron-safety-guard-8b-v3", - "name": "llama-3.1-nemotron-safety-guard-8b-v3", + "cohere/command-r-08-2024": { + "id": "cohere/command-r-08-2024", + "name": "Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] @@ -81207,3270 +130971,3602 @@ "open_weights": true, "limit": { "context": 128000, - "output": 4096 + "output": 4000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.6 } }, - "nvidia/nv-embed-v1": { - "id": "nvidia/nv-embed-v1", - "name": "nv-embed-v1", + "cohere/command-r7b-12-2024": { + "id": "cohere/command-r7b-12-2024", + "name": "Command R7B", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-06-07", - "last_updated": "2025-07-22", + "structured_output": true, + "temperature": true, + "knowledge": "2024-06-01", + "release_date": "2024-12-02", + "last_updated": "2024-12-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 2048 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.0375, + "output": 0.15 } }, - "nvidia/nemotron-3-nano-30b-a3b": { - "id": "nvidia/nemotron-3-nano-30b-a3b", - "name": "nemotron-3-nano-30b-a3b", - "family": "nemotron", + "cohere/command-r-plus-08-2024": { + "id": "cohere/command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-12", - "last_updated": "2024-12", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 10 } }, - "nvidia/gliner-pii": { - "id": "nvidia/gliner-pii", - "name": "gliner-pii", + "cohere/north-mini-code:free": { + "id": "cohere/north-mini-code:free", + "name": "North Mini Code (free)", + "description": "Cohere coding model for practical software engineering and agentic edits", + "family": "north", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2026-06-17", + "last_updated": "2026-06-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 256000, + "output": 64000 }, "cost": { "input": 0, "output": 0 } }, - "nvidia/bevformer": { - "id": "nvidia/bevformer", - "name": "bevformer", + "~x-ai/grok-latest": { + "id": "~x-ai/grok-latest", + "name": "Grok Latest", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-03-18", - "last_updated": "2025-07-20", + "release_date": "2026-07-08", + "last_updated": "2026-07-08", "modalities": { - "input": ["video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 500000, + "output": 1000000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 6, + "cache_read": 0.5 } }, - "nvidia/nemotron-mini-4b-instruct": { - "id": "nvidia/nemotron-mini-4b-instruct", - "name": "nemotron-mini-4b-instruct", + "gryphe/mythomax-l2-13b": { + "id": "gryphe/mythomax-l2-13b", + "name": "MythoMax 13B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2024-08-21", - "last_updated": "2024-08-26", + "knowledge": "2023-06-30", + "release_date": "2023-07-02", + "last_updated": "2023-07-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.06, + "output": 0.06 } }, - "nvidia/cosmos-predict1-5b": { - "id": "nvidia/cosmos-predict1-5b", - "name": "cosmos-predict1-5b", + "stepfun/step-3.7-flash": { + "id": "stepfun/step-3.7-flash", + "name": "Step 3.7 Flash", + "description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-03-18", - "last_updated": "2025-03-18", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2026-01-01", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { "input": ["text", "image", "video"], - "output": ["video"] + "output": ["text"] }, "open_weights": true, "limit": { - "context": 0, - "output": 4096 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "nvidia/llama-3_3-nemotron-super-49b-v1": { - "id": "nvidia/llama-3_3-nemotron-super-49b-v1", - "name": "Llama 3.3 Nemotron Super 49B v1", - "family": "nemotron", + "stepfun/step-3.5-flash": { + "id": "stepfun/step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash lane for quick multimodal reasoning and coding assistance", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-04-07", - "last_updated": "2025-04-07", + "knowledge": "2025-01", + "release_date": "2026-01-29", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.3 } }, - "nvidia/nvidia-nemotron-nano-9b-v2": { - "id": "nvidia/nvidia-nemotron-nano-9b-v2", - "name": "nvidia-nemotron-nano-9b-v2", - "family": "nemotron", - "attachment": false, + "nex-agi/nex-n2-mini": { + "id": "nex-agi/nex-n2-mini", + "name": "Nex-N2-Mini", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "agi", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-08-18", - "last_updated": "2025-08-18", + "release_date": "2026-06-24", + "last_updated": "2026-06-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.025, + "output": 0.1, + "cache_read": 0.0025 } }, - "nvidia/cosmos-transfer2_5-2b": { - "id": "nvidia/cosmos-transfer2_5-2b", - "name": "cosmos-transfer2.5-2b", + "nex-agi/nex-n2-pro": { + "id": "nex-agi/nex-n2-pro", + "name": "Nex-N2-Pro", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "agi", "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "release_date": "2026-06-08", + "last_updated": "2026-06-08", "modalities": { - "input": ["text", "image", "video"], - "output": ["video"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": true, "limit": { - "context": 0, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 1, + "cache_read": 0.025 } }, - "nvidia/nemotron-content-safety-reasoning-4b": { - "id": "nvidia/nemotron-content-safety-reasoning-4b", - "name": "nemotron-content-safety-reasoning-4b", + "undi95/remm-slerp-l2-13b": { + "id": "undi95/remm-slerp-l2-13b", + "name": "ReMM SLERP 13B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "structured_output": true, + "temperature": true, + "knowledge": "2023-06-30", + "release_date": "2023-07-22", + "last_updated": "2023-07-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, + "context": 6144, "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.45, + "output": 0.65 } }, - "nvidia/active-speaker-detection": { - "id": "nvidia/active-speaker-detection", - "name": "Active Speaker Detection", + "~openai/gpt-mini-latest": { + "id": "~openai/gpt-mini-latest", + "name": "OpenAI GPT Mini Latest", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "knowledge": "2025-08-31", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["video"], + "input": ["pdf", "image", "text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 4096 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "nvidia/magpie-tts-zeroshot": { - "id": "nvidia/magpie-tts-zeroshot", - "name": "magpie-tts-zeroshot", + "~openai/gpt-latest": { + "id": "~openai/gpt-latest", + "name": "OpenAI GPT Latest", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": false, - "release_date": "2025-05-22", - "last_updated": "2025-06-12", + "knowledge": "2026-02-16", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "audio"], - "output": ["audio"] + "input": ["pdf", "image", "text"], + "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 4096 + "context": 1050000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "nvidia/streampetr": { - "id": "nvidia/streampetr", - "name": "streampetr", + "~moonshotai/kimi-latest": { + "id": "~moonshotai/kimi-latest", + "name": "MoonshotAI Kimi Latest", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.66, + "output": 3.41, + "cache_read": 0.15 } }, - "nvidia/sparsedrive": { - "id": "nvidia/sparsedrive", - "name": "sparsedrive", - "attachment": true, + "relace/relace-search": { + "id": "relace/relace-search", + "name": "Relace Search", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2025-03-18", - "last_updated": "2025-07-20", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 3 } }, - "nvidia/nemotron-3-content-safety": { - "id": "nvidia/nemotron-3-content-safety", - "name": "nemotron-3-content-safety", + "relace/relace-apply-3": { + "id": "relace/relace-apply-3", + "name": "Relace Apply 3", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, "tool_call": false, + "structured_output": false, "temperature": false, - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "release_date": "2025-09-26", + "last_updated": "2025-09-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.85, + "output": 1.25 } }, - "nvidia/llama-nemotron-embed-vl-1b-v2": { - "id": "nvidia/llama-nemotron-embed-vl-1b-v2", - "name": "llama-nemotron-embed-vl-1b-v2", - "attachment": true, + "ai21/jamba-large-1.7": { + "id": "ai21/jamba-large-1.7", + "name": "Jamba Large 1.7", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "jamba", + "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-02-10", - "last_updated": "2026-02-10", + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 2048 + "context": 256000, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 8 } }, - "nvidia/studiovoice": { - "id": "nvidia/studiovoice", - "name": "studiovoice", + "arcee-ai/coder-large": { + "id": "arcee-ai/coder-large", + "name": "Coder Large", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": false, "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2024-10-03", - "last_updated": "2025-06-13", + "knowledge": "2025-03-31", + "release_date": "2025-05-05", + "last_updated": "2025-05-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 0.8 } }, - "nvidia/llama-nemotron-rerank-vl-1b-v2": { - "id": "nvidia/llama-nemotron-rerank-vl-1b-v2", - "name": "llama-nemotron-rerank-vl-1b-v2", - "attachment": true, + "arcee-ai/virtuoso-large": { + "id": "arcee-ai/virtuoso-large", + "name": "Virtuoso Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-03-31", - "last_updated": "2026-03-31", + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-05", + "last_updated": "2025-05-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 1.2 } }, - "deepseek-ai/deepseek-v3.2": { - "id": "deepseek-ai/deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "arcee-ai/trinity-large-thinking": { + "id": "arcee-ai/trinity-large-thinking", + "name": "Trinity Large Thinking", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "trinity", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 65536 + "context": 262144, + "output": 80000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 0.8, + "cache_read": 0.06 } }, - "deepseek-ai/deepseek-v3.1-terminus": { - "id": "deepseek-ai/deepseek-v3.1-terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", + "mancer/weaver": { + "id": "mancer/weaver", + "name": "Weaver (alpha)", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "alpha", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "knowledge": "2023-06-30", + "release_date": "2023-08-02", + "last_updated": "2023-08-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 8000, + "output": 2000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 1 } }, - "openai/whisper-large-v3": { - "id": "openai/whisper-large-v3", - "name": "Whisper Large v3", - "family": "whisper", - "attachment": false, - "reasoning": false, + "perplexity/sonar-reasoning-pro": { + "id": "perplexity/sonar-reasoning-pro", + "name": "Sonar Reasoning Pro", + "description": "Web-grounded reasoning model for multi-step research and cited answers", + "family": "sonar-reasoning", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2023-09-01", - "last_updated": "2025-09-05", + "structured_output": false, + "temperature": true, + "release_date": "2025-03-07", + "last_updated": "2025-03-07", "modalities": { - "input": ["audio"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 4096 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 8 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "perplexity/sonar": { + "id": "perplexity/sonar", + "name": "Sonar", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-01-27", + "last_updated": "2025-01-27", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 127072, + "output": 127072 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 1 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT-OSS-120B", - "family": "gpt-oss", + "perplexity/sonar-pro": { + "id": "perplexity/sonar-pro", + "name": "Sonar Pro", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "family": "sonar-pro", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-04", - "last_updated": "2025-08-14", + "release_date": "2025-03-07", + "last_updated": "2025-03-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 200000, + "output": 8000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15 } }, - "minimaxai/minimax-m2.7": { - "id": "minimaxai/minimax-m2.7", - "name": "MiniMax-M2.7", - "family": "minimax", - "attachment": false, + "perplexity/sonar-pro-search": { + "id": "perplexity/sonar-pro-search", + "name": "Sonar Pro Search", + "description": "Advanced Sonar search model for deeper research and cited synthesis", + "family": "sonar-pro", + "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-04-11", + "release_date": "2025-10-30", + "last_updated": "2025-10-30", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "output": 8000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15 } }, - "minimaxai/minimax-m2.5": { - "id": "minimaxai/minimax-m2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "perplexity/sonar-deep-research": { + "id": "perplexity/sonar-deep-research", + "name": "Sonar Deep Research", + "description": "Sonar search model for current answers, retrieval, and citation-backed chat", + "family": "sonar-deep-research", "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2025-08", - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2025-03-07", + "last_updated": "2025-03-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 8, + "reasoning": 3 } }, - "z-ai/glm4.7": { - "id": "z-ai/glm4.7", - "name": "GLM-4.7", - "family": "glm", + "openrouter/bodybuilder": { + "id": "openrouter/bodybuilder", + "name": "Body Builder (beta)", + "description": "Preview model for early access evaluation, prototyping, and compatibility testing", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": false, + "release_date": "2025-12-05", + "last_updated": "2025-12-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 0, - "output": 0 + "context": 128000, + "output": 128000 } }, - "z-ai/glm-5.1": { - "id": "z-ai/glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "openrouter/free": { + "id": "openrouter/free", + "name": "Free Models Router", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "release_date": "2026-02-01", + "last_updated": "2026-02-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 200000, + "input": 200000, + "output": 8000 }, "cost": { "input": 0, "output": 0 } }, - "meta/esm2-650m": { - "id": "meta/esm2-650m", - "name": "esm2-650m", + "openrouter/fusion": { + "id": "openrouter/fusion", + "name": "Fusion", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2024-08-29", - "last_updated": "2025-03-10", + "structured_output": false, + "temperature": false, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 - }, - "cost": { - "input": 0, - "output": 0 + "context": 1000000, + "output": 128000 } }, - "meta/llama-3.3-70b-instruct": { - "id": "meta/llama-3.3-70b-instruct", - "name": "Llama 3.3 70b Instruct", + "openrouter/pareto-code": { + "id": "openrouter/pareto-code", + "name": "Pareto Code Router", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 200000 + } + }, + "openrouter/auto": { + "id": "openrouter/auto", + "name": "Auto Router", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "auto", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-11-26", - "last_updated": "2024-11-26", + "release_date": "2023-11-08", + "last_updated": "2023-11-08", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "pdf", "video"], + "output": ["text", "image"] + }, + "open_weights": false, + "limit": { + "context": 2000000, + "output": 2000000 + } + }, + "qwen/qwen3.5-plus-20260420": { + "id": "qwen/qwen3.5-plus-20260420", + "name": "Qwen3.5 Plus 2026-04-20", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", + "modalities": { + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.8, + "cache_write": 0.375 } }, - "meta/esmfold": { - "id": "meta/esmfold", - "name": "esmfold", + "qwen/qwen3-next-80b-a3b-instruct:free": { + "id": "qwen/qwen3-next-80b-a3b-instruct:free", + "name": "Qwen3 Next 80B A3B Instruct (free)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-03-15", - "last_updated": "2025-06-12", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { "input": 0, "output": 0 } }, - "meta/llama-3.2-90b-vision-instruct": { - "id": "meta/llama-3.2-90b-vision-instruct", - "name": "Llama-3.2-90B-Vision-Instruct", - "family": "llama", + "qwen/qwen3-vl-235b-a22b-thinking": { + "id": "qwen/qwen3-vl-235b-a22b-thinking", + "name": "Qwen3 VL 235B A22B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2025-03-31", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.26, + "output": 2.6 } }, - "meta/llama-3.2-11b-vision-instruct": { - "id": "meta/llama-3.2-11b-vision-instruct", - "name": "Llama 3.2 11b Vision Instruct", + "qwen/qwen3-vl-30b-a3b-thinking": { + "id": "qwen/qwen3-vl-30b-a3b-thinking", + "name": "Qwen3 VL 30B A3B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "knowledge": "2025-03-31", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.13, + "output": 1.56 } }, - "meta/llama-guard-4-12b": { - "id": "meta/llama-guard-4-12b", - "name": "Llama Guard 4 12B", - "family": "llama", - "attachment": true, + "qwen/qwen3-coder-plus": { + "id": "qwen/qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Hosted Qwen coder for software agents, repo edits, and long-context code", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2026-04-30", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.65, + "output": 3.25, + "cache_read": 0.13, + "cache_write": 0.8125 } }, - "meta/llama-3.1-70b-instruct": { - "id": "meta/llama-3.1-70b-instruct", - "name": "Llama 3.1 70b Instruct", + "qwen/qwen-plus": { + "id": "qwen/qwen-plus", + "name": "Qwen Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-07-16", - "last_updated": "2024-07-16", + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-09-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.26, + "output": 0.78, + "cache_read": 0.052, + "cache_write": 0.325 } }, - "meta/llama-3.2-1b-instruct": { - "id": "meta/llama-3.2-1b-instruct", - "name": "Llama 3.2 1b Instruct", + "qwen/qwen3-coder-30b-a3b-instruct": { + "id": "qwen/qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 160000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.07, + "output": 0.27 } }, - "meta/llama-4-maverick-17b-128e-instruct": { - "id": "meta/llama-4-maverick-17b-128e-instruct", - "name": "Llama 4 Maverick 17b 128e Instruct", - "attachment": true, - "reasoning": false, + "qwen/qwen3-32b": { + "id": "qwen/qwen3-32b", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-02", - "release_date": "2025-04-01", - "last_updated": "2025-04-01", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.08, + "output": 0.28 } }, - "meta/llama-3.2-3b-instruct": { - "id": "meta/llama-3.2-3b-instruct", - "name": "Llama 3.2 3B Instruct", - "family": "llama", + "qwen/qwen3-next-80b-a3b-instruct": { + "id": "qwen/qwen3-next-80b-a3b-instruct", + "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32000 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.09, + "output": 1.1 } }, - "meta/llama-3.1-8b-instruct": { - "id": "meta/llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", - "attachment": false, + "qwen/qwen3-vl-8b-instruct": { + "id": "qwen/qwen3-vl-8b-instruct", + "name": "Qwen3 VL 8B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2025-10-14", + "last_updated": "2025-10-14", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4096 + "context": 256000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.117, + "output": 0.455 } }, - "qwen/qwen-image-edit": { - "id": "qwen/qwen-image-edit", - "name": "Qwen Image Edit", + "qwen/qwen3.7-plus": { + "id": "qwen/qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { "input": ["text", "image"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.32, + "output": 1.28, + "cache_read": 0.064, + "cache_write": 0.4 } }, - "qwen/qwen3.5-122b-a10b": { - "id": "qwen/qwen3.5-122b-a10b", - "name": "Qwen3.5 122B-A10B", + "qwen/qwen3.6-35b-a3b": { + "id": "qwen/qwen3.6-35b-a3b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-23", - "last_updated": "2026-02-23", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 65536 + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.14, + "output": 1 } }, - "qwen/qwen2.5-coder-32b-instruct": { - "id": "qwen/qwen2.5-coder-32b-instruct", - "name": "Qwen2.5 Coder 32b Instruct", + "qwen/qwen3.7-max": { + "id": "qwen/qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 262144 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-11-06", - "last_updated": "2024-11-06", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 3.75, + "cache_read": 0.25, + "cache_write": 1.5625 } }, - "qwen/qwen3.5-397b-a17b": { - "id": "qwen/qwen3.5-397b-a17b", - "name": "Qwen3.5-397B-A17B", + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", "family": "qwen", - "attachment": true, - "reasoning": true, + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 8192 + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.78, + "output": 3.9, + "cache_read": 0.156, + "cache_write": 0.975 } }, - "qwen/qwen3-next-80b-a3b-thinking": { - "id": "qwen/qwen3-next-80b-a3b-thinking", - "name": "Qwen3-Next-80B-A3B-Thinking", + "qwen/qwen3-8b": { + "id": "qwen/qwen3-8b", + "name": "Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2025-09-05", + "knowledge": "2025-03-31", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.117, + "output": 0.455 } }, - "qwen/qwen3-next-80b-a3b-instruct": { - "id": "qwen/qwen3-next-80b-a3b-instruct", - "name": "Qwen3-Next-80B-A3B-Instruct", + "qwen/qwen-plus-2025-07-28": { + "id": "qwen/qwen-plus-2025-07-28", + "name": "Qwen Plus 0728", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2025-09-05", + "knowledge": "2025-03-31", + "release_date": "2025-09-08", + "last_updated": "2025-09-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 16384 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.26, + "output": 0.78 } }, - "qwen/qwen-image": { - "id": "qwen/qwen-image", - "name": "Qwen Image", + "qwen/qwen3.5-flash-02-23": { + "id": "qwen/qwen3.5-flash-02-23", + "name": "Qwen3.5-Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", "attachment": true, - "reasoning": false, - "tool_call": false, - "structured_output": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2026-02-25", + "last_updated": "2026-02-25", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text", "image", "video"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.065, + "output": 0.26 } }, - "qwen/qwen3-coder-480b-a35b-instruct": { - "id": "qwen/qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", + "qwen/qwen3-coder:free": { + "id": "qwen/qwen3-coder:free", + "name": "Qwen3 Coder 480B A35B (free)", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-04", + "knowledge": "2025-06-30", "release_date": "2025-07-23", "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 66536 + "context": 1048576, + "output": 262000 }, "cost": { "input": 0, "output": 0 } }, - "google/gemma-2-2b-it": { - "id": "google/gemma-2-2b-it", - "name": "Gemma 2 2b It", + "qwen/qwen3-30b-a3b-instruct-2507": { + "id": "qwen/qwen3-30b-a3b-instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2024-07-16", - "last_updated": "2024-07-16", + "knowledge": "2025-06-30", + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.04815, + "output": 0.19305 } }, - "google/gemma-3n-e4b-it": { - "id": "google/gemma-3n-e4b-it", - "name": "Gemma 3n E4b It", - "attachment": true, + "qwen/qwen-2.5-coder-32b-instruct": { + "id": "qwen/qwen-2.5-coder-32b-instruct", + "name": "Qwen2.5 Coder 32B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-06-03", - "last_updated": "2025-06-03", + "knowledge": "2024-06-30", + "release_date": "2024-11-11", + "last_updated": "2024-11-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { "context": 128000, - "output": 4096 + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.66, + "output": 1 } }, - "google/gemma-4-31b-it": { - "id": "google/gemma-4-31b-it", - "name": "Gemma-4-31B-IT", - "family": "gemma", - "attachment": true, + "qwen/qwen3-next-80b-a3b-thinking": { + "id": "qwen/qwen3-next-80b-a3b-thinking", + "name": "Qwen3-Next 80B-A3B (Thinking)", + "description": "Efficient Qwen thinking model for local reasoning, math, and coding agents", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.0975, + "output": 0.78 } }, - "google/gemma-3n-e2b-it": { - "id": "google/gemma-3n-e2b-it", - "name": "Gemma 3n E2b It", - "attachment": true, - "reasoning": false, + "qwen/qwen3-235b-a22b-thinking-2507": { + "id": "qwen/qwen3-235b-a22b-thinking-2507", + "name": "Qwen3 235B A22B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-06-12", - "last_updated": "2025-06-12", + "knowledge": "2025-06-30", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1495, + "output": 1.495 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Gemma-3-27B-IT", - "family": "gemma", + "qwen/qwen3-vl-32b-instruct": { + "id": "qwen/qwen3-vl-32b-instruct", + "name": "Qwen3 VL 32B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2024-12-01", - "last_updated": "2025-09-05", + "release_date": "2025-10-23", + "last_updated": "2025-10-23", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.104, + "output": 0.416 } }, - "google/google-paligemma": { - "id": "google/google-paligemma", - "name": "paligemma", - "attachment": true, + "qwen/qwen3-coder": { + "id": "qwen/qwen3-coder", + "name": "Qwen3 Coder 480B A35B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-05-14", - "last_updated": "2024-08-26", + "knowledge": "2025-06-30", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.22, + "output": 1.8 } }, - "moonshotai/kimi-k2-instruct": { - "id": "moonshotai/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", - "attachment": false, + "qwen/qwen3.6-flash": { + "id": "qwen/qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-01", - "release_date": "2025-01-01", - "last_updated": "2025-09-05", + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1875, + "output": 1.125, + "cache_write": 0.234375 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "qwen/qwen3.5-plus-02-15": { + "id": "qwen/qwen3.5-plus-02-15", + "name": "Qwen3.5 Plus 2026-02-15", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.26, + "output": 1.56 } }, - "moonshotai/kimi-k2-instruct-0905": { - "id": "moonshotai/kimi-k2-instruct-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "qwen/qwen-2.5-7b-instruct": { + "id": "qwen/qwen-2.5-7b-instruct", + "name": "Qwen2.5 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "knowledge": "2024-06-30", + "release_date": "2024-10-16", + "last_updated": "2024-10-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.04, + "output": 0.1 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "qwen/qwen3-vl-8b-thinking": { + "id": "qwen/qwen3-vl-8b-thinking", + "name": "Qwen3 VL 8B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": true, "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11", - "last_updated": "2025-12", + "release_date": "2025-10-14", + "last_updated": "2025-10-14", "modalities": { - "input": ["text"], + "input": ["image", "text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 256000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.117, + "output": 1.365 } }, - "abacusai/dracarys-llama-3_1-70b-instruct": { - "id": "abacusai/dracarys-llama-3_1-70b-instruct", - "name": "dracarys-llama-3.1-70b-instruct", + "qwen/qwen3-max-thinking": { + "id": "qwen/qwen3-max-thinking", + "name": "Qwen3 Max Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-09-11", - "last_updated": "2025-05-22", + "release_date": "2026-02-09", + "last_updated": "2026-02-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.78, + "output": 3.9 } }, - "baai/bge-m3": { - "id": "baai/bge-m3", - "name": "BGE M3", - "family": "bge", + "qwen/qwen3-30b-a3b-thinking-2507": { + "id": "qwen/qwen3-30b-a3b-thinking-2507", + "name": "Qwen3 30B A3B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-01-30", - "last_updated": "2026-04-30", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "temperature": true, + "knowledge": "2025-06-30", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 1024 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.13, + "output": 1.56 } }, - "bytedance/seed-oss-36b-instruct": { - "id": "bytedance/seed-oss-36b-instruct", - "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", - "family": "seed", - "attachment": false, + "qwen/qwen2.5-vl-72b-instruct": { + "id": "qwen/qwen2.5-vl-72b-instruct", + "name": "Qwen2.5 VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": true, "temperature": true, - "release_date": "2025-09-04", - "last_updated": "2025-11-25", + "knowledge": "2024-06-30", + "release_date": "2025-02-01", + "last_updated": "2025-02-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 131072, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 0.75 } }, - "deepseek-ai/deepseek-v4-flash": { - "id": "deepseek-ai/deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", - "attachment": false, + "qwen/qwen3.5-27b": { + "id": "qwen/qwen3.5-27b", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.195, + "output": 1.56 } }, - "deepseek-ai/deepseek-v4-pro": { - "id": "deepseek-ai/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "qwen/qwen3-235b-a22b": { + "id": "qwen/qwen3-235b-a22b", + "name": "Qwen3 235B-A22B", + "description": "Large open Qwen MoE for multilingual reasoning, coding, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 38912 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 131072, + "output": 8192 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0.455, + "output": 1.82 } - } - } - }, - "inference": { - "id": "inference", - "env": ["INFERENCE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://inference.net/v1", - "name": "Inference", - "doc": "https://inference.net/models", - "models": { - "mistral/mistral-nemo-12b-instruct": { - "id": "mistral/mistral-nemo-12b-instruct", - "name": "Mistral Nemo 12B Instruct", - "family": "mistral-nemo", + }, + "qwen/qwen-2.5-72b-instruct": { + "id": "qwen/qwen-2.5-72b-instruct", + "name": "Qwen2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2024-06-30", + "release_date": "2024-09-19", + "last_updated": "2024-09-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4096 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.038, - "output": 0.1 + "input": 0.36, + "output": 0.4 } }, - "meta/llama-3.2-11b-vision-instruct": { - "id": "meta/llama-3.2-11b-vision-instruct", - "name": "Llama 3.2 11B Vision Instruct", - "family": "llama", - "attachment": true, - "reasoning": false, + "qwen/qwen-plus-2025-07-28:thinking": { + "id": "qwen/qwen-plus-2025-07-28:thinking", + "name": "Qwen Plus 0728 (thinking)", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-03-31", + "release_date": "2025-09-08", + "last_updated": "2025-09-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16000, - "output": 4096 + "context": 1000000, + "output": 32768 }, "cost": { - "input": 0.055, - "output": 0.055 + "input": 0.26, + "output": 0.78, + "cache_write": 0.325 } }, - "meta/llama-3.2-1b-instruct": { - "id": "meta/llama-3.2-1b-instruct", - "name": "Llama 3.2 1B Instruct", - "family": "llama", + "qwen/qwen3-coder-next": { + "id": "qwen/qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-02-04", + "last_updated": "2026-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.01, - "output": 0.01 + "input": 0.11, + "output": 0.8, + "cache_read": 0.07 } }, - "meta/llama-3.2-3b-instruct": { - "id": "meta/llama-3.2-3b-instruct", - "name": "Llama 3.2 3B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "qwen/qwen3.6-27b": { + "id": "qwen/qwen3.6-27b", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4096 + "context": 262144, + "output": 262140 }, "cost": { - "input": 0.02, - "output": 0.02 + "input": 0.285, + "output": 2.4, + "cache_read": 0.15 } }, - "meta/llama-3.1-8b-instruct": { - "id": "meta/llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "qwen/qwen3.5-35b-a3b": { + "id": "qwen/qwen3.5-35b-a3b", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 16000, - "output": 4096 + "context": 262144, + "output": 81920 }, "cost": { - "input": 0.025, - "output": 0.025 + "input": 0.14, + "output": 1, + "cache_read": 0.05 } }, - "qwen/qwen-2.5-7b-vision-instruct": { - "id": "qwen/qwen-2.5-7b-vision-instruct", - "name": "Qwen 2.5 7B Vision Instruct", + "qwen/qwen3.5-9b": { + "id": "qwen/qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 125000, - "output": 4096 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.1, + "output": 0.15 } }, - "qwen/qwen3-embedding-4b": { - "id": "qwen/qwen3-embedding-4b", - "name": "Qwen 3 Embedding 4B", + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 2048 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.01, - "output": 0 + "input": 0.385, + "output": 2.45, + "cache_read": 0.111 } }, - "google/gemma-3": { - "id": "google/gemma-3", - "name": "Google Gemma 3", - "family": "gemma", + "qwen/qwen3-vl-30b-a3b-instruct": { + "id": "qwen/qwen3-vl-30b-a3b-instruct", + "name": "Qwen3 VL 30B A3B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-03-31", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 125000, - "output": 4096 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.3 + "input": 0.13, + "output": 0.52 } }, - "osmosis/osmosis-structure-0.6b": { - "id": "osmosis/osmosis-structure-0.6b", - "name": "Osmosis Structure 0.6B", - "family": "osmosis", + "qwen/qwen3-235b-a22b-2507": { + "id": "qwen/qwen3-235b-a22b-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-06-30", + "release_date": "2025-07-21", + "last_updated": "2025-07-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, - "limit": { - "context": 4000, - "output": 2048 + "limit": { + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.09, + "output": 0.55 } - } - } - }, - "inception": { - "id": "inception", - "env": ["INCEPTION_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.inceptionlabs.ai/v1/", - "name": "Inception", - "doc": "https://platform.inceptionlabs.ai/docs", - "models": { - "mercury-edit-2": { - "id": "mercury-edit-2", - "name": "Mercury Edit 2", + }, + "qwen/qwen3-coder-flash": { + "id": "qwen/qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, + "structured_output": false, "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 0.195, + "output": 0.975, + "cache_read": 0.039, + "cache_write": 0.24375 } }, - "mercury-2": { - "id": "mercury-2", - "name": "Mercury 2", - "family": "mercury", + "qwen/qwen3-14b": { + "id": "qwen/qwen3-14b", + "name": "Qwen3 14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01-01", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "knowledge": "2025-03-31", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 50000 + "context": 131702, + "output": 40960 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 0.1, + "output": 0.24 } - } - } - }, - "openai": { - "id": "openai", - "env": ["OPENAI_API_KEY"], - "npm": "@ai-sdk/openai", - "name": "OpenAI", - "doc": "https://platform.openai.com/docs/models", - "models": { - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", - "family": "gpt-codex", + }, + "qwen/qwen3-vl-235b-a22b-instruct": { + "id": "qwen/qwen3-vl-235b-a22b-instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.2, + "output": 0.88, + "cache_read": 0.11 } }, - "gpt-4o-2024-05-13": { - "id": "gpt-4o-2024-05-13", - "name": "GPT-4o (2024-05-13)", - "family": "gpt", - "attachment": true, - "reasoning": false, + "qwen/qwen3-30b-a3b": { + "id": "qwen/qwen3-30b-a3b", + "name": "Qwen3 30B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "knowledge": "2025-03-31", + "release_date": "2025-04-28", + "last_updated": "2025-04-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 16384 }, "cost": { - "input": 5, - "output": 15 + "input": 0.12, + "output": 0.5 } }, - "o1-mini": { - "id": "o1-mini", - "name": "o1-mini", - "family": "o-mini", + "qwen/qwen3.6-max-preview": { + "id": "qwen/qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 131072 + } + ], + "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 262144, "output": 65536 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 1.04, + "output": 6.24, + "cache_write": 1.3 } }, - "gpt-5.2-pro": { - "id": "gpt-5.2-pro", - "name": "GPT-5.2 Pro", - "family": "gpt-pro", + "qwen/qwen3.5-122b-a10b": { + "id": "qwen/qwen3.5-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 21, - "output": 168 + "input": 0.26, + "output": 2.08 } }, - "text-embedding-3-large": { - "id": "text-embedding-3-large", - "name": "text-embedding-3-large", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-01", - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "qwen/qwen3.6-plus": { + "id": "qwen/qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1, + "max": 81920 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 3072 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.13, - "output": 0 + "input": 0.325, + "output": 1.95, + "cache_write": 0.40625 } }, - "gpt-5.3-chat-latest": { - "id": "gpt-5.3-chat-latest", - "name": "GPT-5.3 Chat (latest)", - "family": "gpt", + "amazon/nova-lite-v1": { + "id": "amazon/nova-lite-v1", + "name": "Nova Lite 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-lite", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "knowledge": "2024-10-31", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 300000, + "output": 5120 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.06, + "output": 0.24 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "amazon/nova-premier-v1": { + "id": "amazon/nova-premier-v1", + "name": "Nova Premier 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "structured_output": false, + "temperature": true, + "release_date": "2025-10-31", + "last_updated": "2025-10-31", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 12.5, - "output": 75, - "cache_read": 1.25 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "context": 1000000, + "output": 32000 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - } + "input": 2.5, + "output": 12.5, + "cache_read": 0.625 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "amazon/nova-pro-v1": { + "id": "amazon/nova-pro-v1", + "name": "Nova Pro 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova-pro", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": false, + "temperature": true, + "knowledge": "2024-10-31", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 300000, + "output": 5120 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.8, + "output": 3.2 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, + "amazon/nova-micro-v1": { + "id": "amazon/nova-micro-v1", + "name": "Nova Micro 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-micro", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": false, + "temperature": true, + "knowledge": "2024-10-31", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "output": 5120 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 0.035, + "output": 0.14 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", + "amazon/nova-2-lite-v1": { + "id": "amazon/nova-2-lite-v1", + "name": "Nova 2 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "nova", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "structured_output": false, + "temperature": true, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1000000, + "output": 65535 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.3, + "output": 2.5 } }, - "gpt-4-turbo": { - "id": "gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", - "attachment": true, - "reasoning": false, + "aion-labs/aion-3.0-mini": { + "id": "aion-labs/aion-3.0-mini", + "name": "Aion-3.0-Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2026-07-07", + "last_updated": "2026-07-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "output": 32768 }, "cost": { - "input": 10, - "output": 30 + "input": 0.7, + "output": 1.4, + "cache_read": 0.18 } }, - "text-embedding-ada-002": { - "id": "text-embedding-ada-002", - "name": "text-embedding-ada-002", - "family": "text-embedding", + "aion-labs/aion-rp-llama-3.1-8b": { + "id": "aion-labs/aion-rp-llama-3.1-8b", + "name": "Aion-RP 1.0 (8B)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2022-12", - "release_date": "2022-12-15", - "last_updated": "2022-12-15", + "structured_output": false, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2025-02-04", + "last_updated": "2025-02-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.8, + "output": 1.6 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, + "aion-labs/aion-3.0": { + "id": "aion-labs/aion-3.0", + "name": "Aion-3.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "structured_output": false, + "temperature": true, + "release_date": "2026-07-07", + "last_updated": "2026-07-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 3, + "output": 6, + "cache_read": 0.75 } }, - "o3-pro": { - "id": "o3-pro", - "name": "o3-pro", - "family": "o-pro", - "attachment": true, + "aion-labs/aion-2.0": { + "id": "aion-labs/aion-2.0", + "name": "Aion-2.0", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-06-10", - "last_updated": "2025-06-10", + "structured_output": false, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 20, - "output": 80 + "input": 0.8, + "output": 1.6, + "cache_read": 0.2 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", - "attachment": true, + "inflection/inflection-3-pi": { + "id": "inflection/inflection-3-pi", + "name": "Inflection 3 Pi", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2024-10-31", + "release_date": "2024-10-11", + "last_updated": "2024-10-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 8000, + "output": 1024 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 2.5, + "output": 10 } }, - "o4-mini-deep-research": { - "id": "o4-mini-deep-research", - "name": "o4-mini-deep-research", - "family": "o-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-06-26", - "last_updated": "2024-06-26", + "inflection/inflection-3-productivity": { + "id": "inflection/inflection-3-productivity", + "name": "Inflection 3 Productivity", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2024-10-31", + "release_date": "2024-10-11", + "last_updated": "2024-10-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 8000, + "output": 1024 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 2.5, + "output": 10 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "sao10k/l3.1-euryale-70b": { + "id": "sao10k/l3.1-euryale-70b", + "name": "Llama 3.1 Euryale 70B v2.2", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-08-28", + "last_updated": "2024-08-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 1.5, - "output": 9, - "cache_read": 0.15 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0.85, + "output": 0.85 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, + "sao10k/l3.3-euryale-70b": { + "id": "sao10k/l3.3-euryale-70b", + "name": "Llama 3.3 Euryale 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-12-18", + "last_updated": "2024-12-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 0.65, + "output": 0.75 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, - "tool_call": true, + "sao10k/l3-lunaris-8b": { + "id": "sao10k/l3-lunaris-8b", + "name": "Llama 3 8B Lunaris", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-08-13", + "last_updated": "2024-08-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 8192, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.04, + "output": 0.05 } }, - "gpt-image-1": { - "id": "gpt-image-1", - "name": "gpt-image-1", - "family": "gpt-image", - "attachment": true, + "sao10k/l3.1-70b-hanami-x1": { + "id": "sao10k/l3.1-70b-hanami-x1", + "name": "Llama 3.1 70B Hanami x1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-04-24", - "last_updated": "2025-04-24", - "modalities": { - "input": ["text", "image"], - "output": ["image"] - }, - "open_weights": false, - "limit": { - "context": 0, - "input": 0, - "output": 0 - } - }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, - "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2025-01-08", + "last_updated": "2025-01-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 16000, + "output": 16000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 3, + "output": 3 } }, - "gpt-5.2-chat-latest": { - "id": "gpt-5.2-chat-latest", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", - "attachment": true, + "upstage/solar-pro-3": { + "id": "upstage/solar-pro-3", + "name": "Solar Pro 3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "solar-pro", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex mini", - "family": "gpt-codex", - "attachment": true, + "allenai/olmo-3-32b-think": { + "id": "allenai/olmo-3-32b-think", + "name": "Olmo 3 32B Think", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "allenai", + "attachment": false, "reasoning": true, - "tool_call": true, + "reasoning_options": [], + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2025-11-21", + "last_updated": "2025-11-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.15, + "output": 0.5 } }, - "o1-preview": { - "id": "o1-preview", - "name": "o1-preview", - "family": "o", + "deepseek/deepseek-r1-0528": { + "id": "deepseek/deepseek-r1-0528", + "name": "R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "knowledge": "2025-03-31", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, + "context": 163840, "output": 32768 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.5, + "output": 2.15, + "cache_read": 0.35 } }, - "gpt-4o-2024-08-06": { - "id": "gpt-4o-2024-08-06", - "name": "GPT-4o (2024-08-06)", - "family": "gpt", - "attachment": true, - "reasoning": false, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-08-06", - "last_updated": "2024-08-06", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.077, + "output": 0.154, + "cache_read": 0.0154 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", - "attachment": true, + "deepseek/deepseek-v3.1-terminus": { + "id": "deepseek/deepseek-v3.1-terminus", + "name": "DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 163840, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, + "input": 0.27, + "output": 0.95, "cache_read": 0.13 } }, - "gpt-image-1-mini": { - "id": "gpt-image-1-mini", - "name": "gpt-image-1-mini", - "family": "gpt-image", - "attachment": true, - "reasoning": false, + "deepseek/deepseek-r1-distill-llama-70b": { + "id": "deepseek/deepseek-r1-distill-llama-70b", + "name": "R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": false, - "release_date": "2025-09-26", - "last_updated": "2025-09-26", + "structured_output": false, + "temperature": true, + "knowledge": "2024-07-31", + "release_date": "2025-01-23", + "last_updated": "2025-01-23", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "input": 0, - "output": 0 + "context": 128000, + "output": 8192 + }, + "cost": { + "input": 0.8, + "output": 0.8 } }, - "o1": { - "id": "o1", - "name": "o1", - "family": "o", - "attachment": true, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 384000 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", - "attachment": true, + "deepseek/deepseek-r1": { + "id": "deepseek/deepseek-r1", + "name": "DeepSeek-R1", + "description": "Classic open reasoning model for transparent math, coding, and deliberate problem solving", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-05-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 163840, + "output": 16000 }, "cost": { - "input": 30, - "output": 180, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 60, - "output": 270 - } + "input": 0.7, + "output": 2.5 } }, - "gpt-3.5-turbo": { - "id": "gpt-3.5-turbo", - "name": "GPT-3.5-turbo", - "family": "gpt", + "deepseek/deepseek-v3.2-exp": { + "id": "deepseek/deepseek-v3.2-exp", + "name": "DeepSeek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "knowledge": "2021-09-01", - "release_date": "2023-03-01", - "last_updated": "2023-11-06", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16385, - "output": 4096 + "context": 163840, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 1.5, - "cache_read": 1.25 + "input": 0.27, + "output": 0.41 } }, - "o3-deep-research": { - "id": "o3-deep-research", - "name": "o3-deep-research", - "family": "o", - "attachment": true, - "reasoning": true, + "deepseek/deepseek-chat-v3-0324": { + "id": "deepseek/deepseek-chat-v3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-06-26", - "last_updated": "2024-06-26", + "structured_output": true, + "temperature": true, + "knowledge": "2024-07-31", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 163840, + "output": 16384 }, "cost": { - "input": 10, - "output": 40, - "cache_read": 2.5 + "input": 0.24, + "output": 0.9, + "cache_read": 0.135 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o-mini", + "deepseek/deepseek-chat": { + "id": "deepseek/deepseek-chat", + "name": "DeepSeek Chat", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "temperature": true, + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 131072, + "output": 16000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.2002, + "output": 0.8001 } }, - "text-embedding-3-small": { - "id": "text-embedding-3-small", - "name": "text-embedding-3-small", - "family": "text-embedding", + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2024-01", - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8191, - "output": 1536 + "context": 131072, + "output": 64000 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.2145, + "output": 0.32175, + "cache_read": 0.02145 } }, - "o1-pro": { - "id": "o1-pro", - "name": "o1-pro", - "family": "o-pro", - "attachment": true, + "deepseek/deepseek-chat-v3.1": { + "id": "deepseek/deepseek-chat-v3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2025-03-19", - "last_updated": "2025-03-19", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 163840, + "output": 32768 }, "cost": { - "input": 150, - "output": 600 + "input": 0.21, + "output": 0.79, + "cache_read": 0.13 } }, - "gpt-4": { - "id": "gpt-4", - "name": "GPT-4", - "family": "gpt", - "attachment": true, - "reasoning": false, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": false, + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 8192 + "context": 204800, + "output": 196608 }, "cost": { - "input": 30, - "output": 60 + "input": 0.15, + "output": 0.9, + "cache_read": 0.05 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "GPT-5-Codex", - "family": "gpt-codex", + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax-M2.1", + "description": "Earlier MiniMax agent model for practical coding and productivity tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "interleaved": { + "field": "reasoning_details" + }, + "structured_output": false, + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + "minimax/minimax-01": { + "id": "minimax/minimax-01", + "name": "MiniMax-01", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2024-03-31", + "release_date": "2025-01-15", + "last_updated": "2025-01-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "context": 1000192, + "output": 1000192 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - } + "input": 0.2, + "output": 1.1 } }, - "gpt-5.1-chat-latest": { - "id": "gpt-5.1-chat-latest", - "name": "GPT-5.1 Chat", - "family": "gpt-codex", + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "gpt-5.3-codex-spark": { - "id": "gpt-5.3-codex-spark", - "name": "GPT-5.3 Codex Spark", - "family": "gpt-codex-spark", - "attachment": true, + "minimax/minimax-m2": { + "id": "minimax/minimax-m2", + "name": "MiniMax-M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_details" + }, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 100000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.255, + "output": 1.02 } }, - "chatgpt-image-latest": { - "id": "chatgpt-image-latest", - "name": "chatgpt-image-latest", - "family": "gpt-image", - "attachment": true, + "minimax/minimax-m2-her": { + "id": "minimax/minimax-m2-her", + "name": "MiniMax M2-her", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "structured_output": false, + "temperature": true, + "release_date": "2026-01-23", + "last_updated": "2026-01-23", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "input": 0, - "output": 0 + "context": 65536, + "output": 2048 + }, + "cost": { + "input": 0.3, + "output": 1.2, + "cache_read": 0.03 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "GPT-4.1 nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": false, + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 204800, + "output": 196608 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.03 + "input": 0.24, + "output": 0.96 } }, - "o3": { - "id": "o3", - "name": "o3", - "family": "o", - "attachment": true, + "minimax/minimax-m1": { + "id": "minimax/minimax-m1", + "name": "MiniMax M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "structured_output": false, + "temperature": true, + "knowledge": "2024-06-30", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "output": 40000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.4, + "output": 2.2 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "kwaipilot/kat-coder-pro-v2": { + "id": "kwaipilot/kat-coder-pro-v2", + "name": "KAT-Coder-Pro V2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "kat-coder", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 272000 + "context": 256000, + "output": 80000 }, "cost": { - "input": 15, - "output": 120 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, + "nousresearch/hermes-3-llama-3.1-405b:free": { + "id": "nousresearch/hermes-3-llama-3.1-405b:free", + "name": "Hermes 3 405B Instruct (free)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "hermes", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "knowledge": "2023-12-31", + "release_date": "2024-08-16", + "last_updated": "2024-08-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0, + "output": 0 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", - "attachment": true, + "nousresearch/hermes-4-405b": { + "id": "nousresearch/hermes-4-405b", + "name": "Hermes 4 405B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "hermes", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1, + "output": 3 } }, - "gpt-5-chat-latest": { - "id": "gpt-5-chat-latest", - "name": "GPT-5 Chat (latest)", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "nousresearch/hermes-3-llama-3.1-70b": { + "id": "nousresearch/hermes-3-llama-3.1-70b", + "name": "Hermes 3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "nousresearch", + "attachment": false, + "reasoning": false, "tool_call": false, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2023-12-31", + "release_date": "2024-08-18", + "last_updated": "2024-08-18", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.7, + "output": 0.7 } }, - "gpt-image-1.5": { - "id": "gpt-image-1.5", - "name": "gpt-image-1.5", - "family": "gpt-image", - "attachment": true, + "nousresearch/hermes-3-llama-3.1-405b": { + "id": "nousresearch/hermes-3-llama-3.1-405b", + "name": "Hermes 3 405B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "nousresearch", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-11-25", - "last_updated": "2025-11-25", + "structured_output": true, + "temperature": true, + "knowledge": "2023-12-31", + "release_date": "2024-08-16", + "last_updated": "2024-08-16", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "input": 0, - "output": 0 + "context": 131072, + "output": 16384 + }, + "cost": { + "input": 1, + "output": 1 } }, - "gpt-5.5-pro": { - "id": "gpt-5.5-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", - "attachment": true, + "nousresearch/hermes-4-70b": { + "id": "nousresearch/hermes-4-70b", + "name": "Hermes 4 70B", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "hermes", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "temperature": true, + "knowledge": "2024-08-31", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 30, - "output": 180, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ], - "context_over_200k": { - "input": 60, - "output": 270 - } + "input": 0.13, + "output": 0.4 } - }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + } + } + }, + "jiekou": { + "id": "jiekou", + "env": ["JIEKOU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.jiekou.ai/openai", + "name": "Jiekou.AI", + "doc": "https://docs.jiekou.ai/docs/support/quickstart?utm_source=github_models.dev", + "models": { + "o3": { + "id": "o3", + "name": "o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 10, + "output": 40 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", + "grok-code-fast-1": { + "id": "grok-code-fast-1", + "name": "grok-code-fast-1", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.18, + "output": 1.35 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "gpt-5.2-pro": { + "id": "gpt-5.2-pro", + "name": "gpt-5.2-pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -84478,251 +134574,254 @@ "open_weights": false, "limit": { "context": 400000, - "input": 272000, "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 18.9, + "output": 151.2 } }, - "gpt-4o-2024-11-20": { - "id": "gpt-4o-2024-11-20", - "name": "GPT-4o (2024-11-20)", - "family": "gpt", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "gemini-2.5-pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-11-20", - "last_updated": "2024-11-20", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 1.125, + "output": 9 } - } - } - }, - "requesty": { - "id": "requesty", - "env": ["REQUESTY_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://router.requesty.ai/v1", - "name": "Requesty", - "doc": "https://requesty.ai/solution/llm-routing/models", - "models": { - "xai/grok-4-fast": { - "id": "xai/grok-4-fast", - "name": "Grok 4 Fast", - "family": "grok", + }, + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "claude-haiku-4-5-20251001", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, + "context": 20000, "output": 64000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05, - "cache_write": 0.2 + "input": 0.9, + "output": 4.5 } }, - "xai/grok-4": { - "id": "xai/grok-4", - "name": "Grok 4", - "family": "grok", + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "gpt-5-pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-09", - "last_updated": "2025-09-09", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 400000, + "output": 272000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75, - "cache_write": 3 + "input": 13.5, + "output": 108 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "GPT-5.1-Codex-Max", - "family": "gpt-codex", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "gemini-2.5-flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 1.1, - "output": 9, - "cache_read": 0.11 + "input": 0.27, + "output": 2.25 } }, - "openai/gpt-5-chat": { - "id": "openai/gpt-5-chat", - "name": "GPT-5 Chat (latest)", - "family": "gpt-codex", + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 1.25, - "output": 10 + "input": 1.1, + "output": 4.4 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT-5.2 Pro", - "family": "gpt-pro", + "gemini-2.5-flash-lite-preview-09-2025": { + "id": "gemini-2.5-flash-lite-preview-09-2025", + "name": "gemini-2.5-flash-lite-preview-09-2025", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 21, - "output": 168 + "input": 0.09, + "output": 0.36 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "claude-opus-4-1-20250805", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 200000, "output": 32000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.03 + "input": 13.5, + "output": 67.5 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", + "grok-4-1-fast-non-reasoning": { + "id": "grok-4-1-fast-non-reasoning", + "name": "grok-4-1-fast-non-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16000, - "output": 4000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.01 + "input": 0.18, + "output": 0.45 } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT-5.3-Codex", - "family": "gpt-codex", + "gpt-5-chat-latest": { + "id": "gpt-5-chat-latest", + "name": "gpt-5-chat-latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, @@ -84731,159 +134830,157 @@ "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.125, + "output": 9 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o Mini", - "family": "gpt-mini", + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "claude-opus-4-5-20251101", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 4.5, + "output": 22.5 } }, - "openai/gpt-5.1-chat": { - "id": "openai/gpt-5.1-chat", - "name": "GPT-5.1 Chat", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "gpt-5.1-codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1.125, + "output": 9 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "o4 Mini", - "family": "o-mini", + "gpt-5.1-codex-max": { + "id": "gpt-5.1-codex-max", + "name": "gpt-5.1-codex-max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 - } - }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "family": "gpt-codex", + "input": 1.125, + "output": 9 + } + }, + "claude-opus-4-20250514": { + "id": "claude-opus-4-20250514", + "name": "claude-opus-4-20250514", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 13.5, + "output": 67.5 } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT-5.1-Codex-Mini", - "family": "gpt-codex", + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 100000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 1.1, + "output": 4.4 } }, - "openai/gpt-5-image": { - "id": "openai/gpt-5-image", - "name": "GPT-5 Image", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "gpt-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-10-14", - "last_updated": "2025-10-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { @@ -84891,107 +134988,110 @@ "output": 128000 }, "cost": { - "input": 5, - "output": 10, - "cache_read": 1.25 + "input": 1.575, + "output": 12.6 } }, - "openai/gpt-5.1": { - "id": "openai/gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "grok-4-0709": { + "id": "grok-4-0709", + "name": "grok-4-0709", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 256000, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 2.7, + "output": 13.5 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "gemini-2.5-flash-lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 30, - "output": 180, - "cache_read": 30 + "input": 0.09, + "output": 0.36 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5 Codex", - "family": "gpt-codex", + "claude-sonnet-4-20250514": { + "id": "claude-sonnet-4-20250514", + "name": "claude-sonnet-4-20250514", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 2.7, + "output": 13.5 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", - "family": "gpt", + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "gpt-5.1-codex-mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "audio", "image", "video"], - "output": ["text", "audio", "image"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { @@ -84999,130 +135099,140 @@ "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.225, + "output": 1.8 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT-4.1 Mini", - "family": "gpt-mini", + "grok-4-fast-reasoning": { + "id": "grok-4-fast-reasoning", + "name": "grok-4-fast-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.18, + "output": 0.45 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1-Codex", - "family": "gpt-codex", + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "claude-opus-4-6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "knowledge": "2025-05-31", + "release_date": "2026-02", + "last_updated": "2026-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 5, + "output": 25 } }, - "google/gemini-3-flash-preview": { - "id": "google/gemini-3-flash-preview", - "name": "Gemini 3 Flash", + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "claude-sonnet-4-5-20250929", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 1 + "input": 2.7, + "output": 13.5 } }, - "google/gemini-3-pro-preview": { - "id": "google/gemini-3-pro-preview", - "name": "Gemini 3 Pro", - "family": "gemini-pro", + "grok-4-1-fast-reasoning": { + "id": "grok-4-1-fast-reasoning", + "name": "grok-4-1-fast-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "cache_write": 4.5 + "input": 0.18, + "output": 0.45 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", + "gemini-3-pro-preview": { + "id": "gemini-3-pro-preview", + "name": "gemini-3-pro-preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, @@ -85131,251 +135241,223 @@ "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "cache_write": 0.55 + "input": 1.8, + "output": 10.8 } }, - "anthropic/claude-haiku-4-5": { - "id": "anthropic/claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "gpt-5-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-01", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 62000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.225, + "output": 1.8 } }, - "anthropic/claude-sonnet-4-6": { - "id": "anthropic/claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "gpt-5-nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 0.045, + "output": 0.36 } }, - "anthropic/claude-3-7-sonnet": { - "id": "anthropic/claude-3-7-sonnet", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", + "grok-4-fast-non-reasoning": { + "id": "grok-4-fast-non-reasoning", + "name": "grok-4-fast-non-reasoning", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-01", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 2000000, + "output": 2000000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.18, + "output": 0.45 } }, - "anthropic/claude-opus-4-5": { - "id": "anthropic/claude-opus-4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "gemini-3-flash-preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.5, + "output": 3 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude Opus 4", - "family": "claude-opus", + "gemini-2.5-flash-preview-05-20": { + "id": "gemini-2.5-flash-preview-05-20", + "name": "gemini-2.5-flash-preview-05-20", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1048576, + "output": 200000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.135, + "output": 3.15 } }, - "anthropic/claude-opus-4-6": { - "id": "anthropic/claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "gemini-2.5-flash-lite-preview-06-17": { + "id": "gemini-2.5-flash-lite-preview-06-17", + "name": "gemini-2.5-flash-lite-preview-06-17", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "video", "image", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 0.09, + "output": 0.36 } }, - "openai/gpt-5.2-chat": { - "id": "openai/gpt-5.2-chat", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", + "gemini-2.5-pro-preview-06-05": { + "id": "gemini-2.5-pro-preview-06-05", + "name": "gemini-2.5-pro-preview-06-05", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 200000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.125, + "output": 9 } }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "gpt-5-codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -85386,67 +135468,50 @@ "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.125, + "output": 9 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "gpt-5.2-codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, + "context": 400000, "output": 128000 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 1.75, + "output": 14 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", + "gpt-5.1": { + "id": "gpt-5.1", + "name": "gpt-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "temperature": true, + "release_date": "2026-02", + "last_updated": "2026-02", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -85454,289 +135519,276 @@ "open_weights": false, "limit": { "context": 400000, - "output": 272000 + "output": 128000 }, "cost": { - "input": 15, - "output": 120 + "input": 1.125, + "output": 9 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "moonshotai/kimi-k2-instruct": { + "id": "moonshotai/kimi-k2-instruct", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1047576, - "output": 32768 + "context": 131072, + "output": 131072 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.57, + "output": 2.3 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 262143 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31, - "cache_write": 2.375, - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - }, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.6, + "output": 3 } }, - "anthropic/claude-opus-4-1": { - "id": "anthropic/claude-opus-4-1", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "moonshotai/kimi-k2-0905": { + "id": "moonshotai/kimi-k2-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.6, + "output": 2.5 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, + "minimaxai/minimax-m1-80k": { + "id": "minimaxai/minimax-m1-80k", + "name": "MiniMax M1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 40000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.55, + "output": 2.2 } }, - "anthropic/claude-sonnet-4-5": { - "id": "anthropic/claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", + "baidu/ernie-4.5-vl-424b-a47b": { + "id": "baidu/ernie-4.5-vl-424b-a47b", + "name": "ERNIE 4.5 VL 424B A47B", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "ernie", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 123000, + "output": 16000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.42, + "output": 1.25 } - } - } - }, - "digitalocean": { - "id": "digitalocean", - "env": ["DIGITALOCEAN_ACCESS_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://inference.do-ai.run/v1", - "name": "DigitalOcean", - "doc": "https://docs.digitalocean.com/products/gradient-ai-platform/details/models/", - "models": { - "openai-gpt-4o-mini": { - "id": "openai-gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", - "attachment": true, + }, + "baidu/ernie-4.5-300b-a47b-paddle": { + "id": "baidu/ernie-4.5-300b-a47b-paddle", + "name": "ERNIE 4.5 300B A47B", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "ernie", + "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 123000, + "output": 12000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.075 + "input": 0.28, + "output": 1.1 } }, - "multi-qa-mpnet-base-dot-v1": { - "id": "multi-qa-mpnet-base-dot-v1", - "name": "Multi-QA-mpnet-base-dot-v1", - "family": "text-embedding", + "xiaomimimo/mimo-v2-flash": { + "id": "xiaomimimo/mimo-v2-flash", + "name": "XiaomiMiMo/MiMo-V2-Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2021-08-30", - "last_updated": "2026-04-16", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 512, - "output": 768 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.009, + "input": 0, "output": 0 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "zai-org/glm-4.7": { + "id": "zai-org/glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": false, - "knowledge": "2025-01", + "temperature": true, "release_date": "2026-01", - "last_updated": "2026-04-16", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.5, - "output": 2.7 + "input": 0.6, + "output": 2.2 } }, - "nemotron-3-nano-omni": { - "id": "nemotron-3-nano-omni", - "name": "Nemotron Nano 3 Omni", - "family": "nemotron", + "zai-org/glm-4.5v": { + "id": "zai-org/glm-4.5v", + "name": "GLM 4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-28", - "last_updated": "2026-04-30", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { "context": 65536, - "output": 65536 + "output": 16384 }, "cost": { - "input": 0.5, - "output": 0.9 + "input": 0.6, + "output": 1.8 } }, - "llama3-8b-instruct": { - "id": "llama3-8b-instruct", - "name": "Llama 3.1 Instruct (8B)", - "family": "llama", + "zai-org/glm-4.5": { + "id": "zai-org/glm-4.5", + "name": "GLM-4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] @@ -85744,2992 +135796,3580 @@ "open_weights": true, "limit": { "context": 131072, - "output": 131072 - }, - "cost": { - "input": 0.198, - "output": 0.198 - } - }, - "anthropic-claude-opus-4.7": { - "id": "anthropic-claude-opus-4.7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 128000 + "output": 98304 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.6, + "output": 2.2 } }, - "anthropic-claude-sonnet-4": { - "id": "anthropic-claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, + "zai-org/glm-4.7-flash": { + "id": "zai-org/glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 64000 - }, - "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75 - } - } - }, - "wan2-2-t2v-a14b": { - "id": "wan2-2-t2v-a14b", - "name": "Wan2.2-T2V-A14B", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-07-28", - "last_updated": "2026-04-30", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], - "output": ["video"] + "output": ["text"] }, "open_weights": true, "limit": { - "context": 100, - "output": 1 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 0 + "input": 0.07, + "output": 0.4 } }, - "qwen-2.5-14b-instruct": { - "id": "qwen-2.5-14b-instruct", - "name": "Qwen 2.5 14B Instruct", + "qwen/qwen3-235b-a22b-instruct-2507": { + "id": "qwen/qwen3-235b-a22b-instruct-2507", + "name": "Qwen3 235B A22B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 131072 - } - }, - "openai-gpt-5.4": { - "id": "openai-gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 16384 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.15, + "output": 0.8 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen 3.5 397B A17B", - "family": "qwen3.5", + "qwen/qwen3-next-80b-a3b-instruct": { + "id": "qwen/qwen3-next-80b-a3b-instruct", + "name": "Qwen3 Next 80B A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-04-30", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 81920 + "context": 65536, + "output": 65536 }, "cost": { - "input": 0.55, - "output": 3.5 + "input": 0.15, + "output": 1.5 } }, - "openai-o3": { - "id": "openai-o3", - "name": "o3", - "family": "o", - "attachment": true, + "qwen/qwen3-next-80b-a3b-thinking": { + "id": "qwen/qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 65536, + "output": 65536 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.15, + "output": 1.5 } }, - "e5-large-v2": { - "id": "e5-large-v2", - "name": "E5 Large v2", - "family": "text-embedding", + "qwen/qwen3-235b-a22b-thinking-2507": { + "id": "qwen/qwen3-235b-a22b-thinking-2507", + "name": "Qwen3 235B A22b Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-05-19", - "last_updated": "2026-04-30", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 512, - "output": 1024 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.3, + "output": 3 } }, - "openai-gpt-5.2-pro": { - "id": "openai-gpt-5.2-pro", - "name": "GPT-5.2 pro", - "family": "gpt-pro", - "attachment": true, + "qwen/qwen3-32b-fp8": { + "id": "qwen/qwen3-32b-fp8", + "name": "Qwen3 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 40960, + "output": 20000 }, "cost": { - "input": 21, - "output": 168 + "input": 0.1, + "output": 0.45 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM 5", - "family": "glm", + "qwen/qwen3-30b-a3b-fp8": { + "id": "qwen/qwen3-30b-a3b-fp8", + "name": "Qwen3 30B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "release_date": "2026-02-11", - "last_updated": "2026-04-16", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 128000 + "context": 40960, + "output": 20000 }, "cost": { - "input": 1, - "output": 3.2 + "input": 0.09, + "output": 0.45 } }, - "openai-gpt-5.4-nano": { - "id": "openai-gpt-5.4-nano", - "name": "GPT-5.4 nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, + "qwen/qwen3-coder-next": { + "id": "qwen/qwen3-coder-next", + "name": "qwen/qwen3-coder-next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "release_date": "2026-02", + "last_updated": "2026-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "output": 1.5 } }, - "mistral-7b-instruct-v0.3": { - "id": "mistral-7b-instruct-v0.3", - "name": "Mistral 7B Instruct v0.3", - "family": "mistral", + "qwen/qwen3-coder-480b-a35b-instruct": { + "id": "qwen/qwen3-coder-480b-a35b-instruct", + "name": "Qwen3 Coder 480B A35B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2024-05-22", - "last_updated": "2024-05-22", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 262144, + "output": 65536 + }, + "cost": { + "input": 0.29, + "output": 1.2 } }, - "llama3.3-70b-instruct": { - "id": "llama3.3-70b-instruct", - "name": "Llama 3.3 Instruct 70B", - "family": "llama", + "qwen/qwen3-235b-a22b-fp8": { + "id": "qwen/qwen3-235b-a22b-fp8", + "name": "Qwen3 235B A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 40960, + "output": 20000 }, "cost": { - "input": 0.65, - "output": 0.65 + "input": 0.2, + "output": 0.8 } }, - "mistral-3-14B": { - "id": "mistral-3-14B", - "name": "Ministral 3 14B Instruct", - "family": "ministral", + "deepseek/deepseek-r1-0528": { + "id": "deepseek/deepseek-r1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-12-15", - "last_updated": "2026-04-30", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 128000 + "context": 163840, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.7, + "output": 2.5 } }, - "deepseek-r1-distill-llama-70b": { - "id": "deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", + "deepseek/deepseek-v3-0324": { + "id": "deepseek/deepseek-v3-0324", + "name": "DeepSeek V3 0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-01-30", - "last_updated": "2025-01-30", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 32768 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.99, - "output": 0.99 + "input": 0.28, + "output": 1.14 } }, - "alibaba-qwen3-32b": { - "id": "alibaba-qwen3-32b", - "name": "Qwen3-32B", - "family": "qwen", + "deepseek/deepseek-v3.1": { + "id": "deepseek/deepseek-v3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32767 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2026-04-16", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131000, - "output": 40960 + "context": 163840, + "output": 32768 }, "cost": { - "input": 0.25, - "output": 0.55 + "input": 0.27, + "output": 1 } }, - "anthropic-claude-opus-4.5": { - "id": "anthropic-claude-opus-4.5", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "Minimax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 131071 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.3, + "output": 1.2 } - }, - "openai-o1": { - "id": "openai-o1", - "name": "o1", - "family": "o", + } + } + }, + "nova": { + "id": "nova", + "env": ["NOVA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.nova.amazon.com/v1", + "name": "Nova", + "doc": "https://nova.amazon.com/dev/documentation", + "models": { + "nova-2-pro-v1": { + "id": "nova-2-pro-v1", + "name": "Nova 2 Pro", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "nova-pro", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "temperature": true, + "release_date": "2025-12-03", + "last_updated": "2026-01-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0, + "output": 0, + "reasoning": 0 } }, - "anthropic-claude-3-opus": { - "id": "anthropic-claude-3-opus", - "name": "Claude 3 Opus", - "family": "claude-opus", + "nova-2-lite-v1": { + "id": "nova-2-lite-v1", + "name": "Nova 2 Lite", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "nova-lite", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-08", - "release_date": "2024-02-29", - "last_updated": "2024-02-29", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 1000000, + "output": 64000 }, - "status": "deprecated", "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0, + "reasoning": 0 } - }, - "stable-diffusion-3.5-large": { - "id": "stable-diffusion-3.5-large", - "name": "Stable Diffusion 3.5 Large", - "family": "stable-diffusion", + } + } + }, + "alibaba-token-plan": { + "id": "alibaba-token-plan", + "env": ["ALIBABA_TOKEN_PLAN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1", + "name": "Alibaba Token Plan", + "doc": "https://www.alibabacloud.com/help/en/model-studio/token-plan-overview", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-10-22", - "last_updated": "2026-04-30", - "modalities": { - "input": ["text"], - "output": ["image"] - }, - "open_weights": true, - "limit": { - "context": 256, - "output": 1 - }, - "cost": { - "input": 0.08, - "output": 0 - } - }, - "openai-gpt-5-nano": { - "id": "openai-gpt-5-nano", - "name": "GPT-5 nano", - "family": "gpt-nano", - "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "llama-4-maverick": { - "id": "llama-4-maverick", - "name": "Llama 4 Maverick 17B 128E Instruct", - "family": "llama", - "attachment": true, - "reasoning": false, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2026-04-30", + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 1000000, - "output": 16384 + "output": 64000 }, "cost": { - "input": 0.25, - "output": 0.87 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "anthropic-claude-4.5-sonnet": { - "id": "anthropic-claude-4.5-sonnet", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 64000 + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75 - } + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "qwen3-embedding-0.6b": { - "id": "qwen3-embedding-0.6b", - "name": "Qwen3 Embedding 0.6B", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": false, - "release_date": "2025-06-03", - "last_updated": "2026-04-16", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8000, - "output": 1024 + "context": 262144, + "output": 262144 }, - "status": "beta", "cost": { - "input": 0.04, - "output": 0 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "anthropic-claude-4.5-haiku": { - "id": "anthropic-claude-4.5-haiku", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 202752, + "output": 128000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 1, - "cache_write": 1.25 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "gte-large-en-v1.5": { - "id": "gte-large-en-v1.5", - "name": "GTE Large (v1.5)", - "family": "text-embedding", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-03-27", - "last_updated": "2026-04-16", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 1024 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.09, - "output": 0 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "openai-gpt-4.1": { - "id": "openai-gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "wan2.7-image-pro": { + "id": "wan2.7-image-pro", + "name": "Wan2.7 Image Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 8192, + "output": 0 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0, + "output": 0 } }, - "anthropic-claude-3.5-haiku": { - "id": "anthropic-claude-3.5-haiku", - "name": "Claude 3.5 Haiku", - "family": "claude-haiku", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-11-05", - "last_updated": "2024-11-05", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 1000000, + "output": 131072 }, - "status": "deprecated", "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "openai-gpt-5.2": { - "id": "openai-gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-3.2": { - "id": "deepseek-3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-12-02", - "last_updated": "2026-04-30", + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 1.6 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "nemotron-3-nano-30b": { - "id": "nemotron-3-nano-30b", - "name": "Nemotron 3 Nano 30B A3B", - "family": "nemotron", + "qwen-image-2.0": { + "id": "qwen-image-2.0", + "name": "Qwen Image 2.0", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 8192, + "output": 0 + }, + "cost": { + "input": 0, + "output": 0 } }, - "anthropic-claude-opus-4": { - "id": "anthropic-claude-opus-4", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 196608, + "input": 196601, + "output": 24576 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "openai-gpt-oss-20b": { - "id": "openai-gpt-oss-20b", - "name": "gpt-oss-20b", - "family": "gpt-oss", - "attachment": false, + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-08-05", - "last_updated": "2026-04-16", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.45 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "qwen3-coder-flash": { - "id": "qwen3-coder-flash", - "name": "Qwen3 Coder Flash", + "qwen-image-2.0-pro": { + "id": "qwen-image-2.0-pro", + "name": "Qwen Image 2.0 Pro", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2026-04-30", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 8192, + "output": 0 }, "cost": { - "input": 0.45, - "output": 1.7 + "input": 0, + "output": 0 } }, - "openai-o3-mini": { - "id": "openai-o3-mini", - "name": "o3-mini", - "family": "o-mini", + "wan2.7-image": { + "id": "wan2.7-image", + "name": "Wan2.7 Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 8192, + "output": 0 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0, + "output": 0 } }, - "openai-gpt-oss-120b": { - "id": "openai-gpt-oss-120b", - "name": "gpt-oss-120b", - "family": "gpt-oss", + "glm-5": { + "id": "glm-5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-08-05", - "last_updated": "2026-04-16", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 131072 - }, - "cost": { - "input": 0.1, - "output": 0.7 - } - }, - "gemma-4-31B-it": { - "id": "gemma-4-31B-it", - "name": "Gemma 4 31B", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-04-30", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 202752, + "output": 16384 }, "cost": { - "input": 0.18, - "output": 0.5 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "nemotron-nano-12b-v2-vl": { - "id": "nemotron-nano-12b-v2-vl", - "name": "Nemotron Nano 12B v2 VL", - "family": "nemotron", - "attachment": true, + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-01", - "last_updated": "2026-04-30", + "knowledge": "2025-01", + "release_date": "2025-12-03", + "last_updated": "2025-12-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0, + "output": 0 } }, - "anthropic-claude-4.1-opus": { - "id": "anthropic-claude-4.1-opus", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 81920 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", - "attachment": true, + } + } + }, + "alibaba": { + "id": "alibaba", + "env": ["DASHSCOPE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1", + "name": "Alibaba", + "doc": "https://www.alibabacloud.com/help/en/model-studio/models", + "models": { + "qwen3-omni-flash": { + "id": "qwen3-omni-flash", + "name": "Qwen3-Omni Flash", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 65536, + "output": 16384 }, "cost": { - "input": 0.95, - "output": 4 + "input": 0.43, + "output": 1.66, + "input_audio": 3.81, + "output_audio": 15.11 } }, - "openai-gpt-image-2": { - "id": "openai-gpt-image-2", - "name": "GPT Image 2", - "family": "gpt-image", - "attachment": true, + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Hosted Qwen coder for software agents, repo edits, and long-context code", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-04-24", - "last_updated": "2025-04-24", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 1048576, + "output": 65536 + }, + "cost": { + "input": 1, + "output": 5 } }, - "anthropic-claude-4.6-sonnet": { - "id": "anthropic-claude-4.6-sonnet", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "qwen-plus": { + "id": "qwen-plus", + "name": "Qwen Plus", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-09-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 64000 + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.3, - "cache_write": 3.75 - } + "input": 0.4, + "output": 1.2, + "reasoning": 4 } }, - "openai-gpt-5-mini": { - "id": "openai-gpt-5-mini", - "name": "GPT-5 mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3-Coder 30B-A3B Instruct", + "description": "Smaller Qwen coder for efficient local agents and repo-level fixes", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.45, + "output": 2.25 } }, - "anthropic-claude-haiku-4.5": { - "id": "anthropic-claude-haiku-4.5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, - "reasoning": true, + "qwen3-omni-flash-realtime": { + "id": "qwen3-omni-flash-realtime", + "name": "Qwen3-Omni Flash Realtime", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2024-04", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 65536, + "output": 16384 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 1, - "cache_write": 1.25 + "input": 0.52, + "output": 1.99, + "input_audio": 4.57, + "output_audio": 18.13 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3 32B", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 393216 + "context": 131072, + "output": 16384 }, "cost": { - "input": 1.74, - "output": 3.48 + "input": 0.7, + "output": 2.8, + "reasoning": 8.4 } }, - "ministral-3-8b-instruct-2512": { - "id": "ministral-3-8b-instruct-2512", - "name": "Ministral 3 8B", - "family": "ministral", - "attachment": true, + "qwen-omni-turbo-realtime": { + "id": "qwen-omni-turbo-realtime", + "name": "Qwen-Omni Turbo Realtime", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-12-15", - "last_updated": "2025-12-15", + "knowledge": "2024-04", + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 32768, + "output": 2048 + }, + "cost": { + "input": 0.27, + "output": 1.07, + "input_audio": 4.44, + "output_audio": 8.89 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax-m2.5", + "qwen-plus-character-ja": { + "id": "qwen-plus-character-ja", + "name": "Qwen Plus Character (Japanese)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2026-02-12", - "last_updated": "2026-04-16", + "knowledge": "2024-04", + "release_date": "2024-01", + "last_updated": "2024-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 128000 + "context": 8192, + "output": 512 }, - "status": "beta", "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.5, + "output": 1.4 } }, - "openai-gpt-image-1": { - "id": "openai-gpt-image-1", - "name": "GPT Image 1", - "family": "gpt-image", - "attachment": true, + "qwen3-next-80b-a3b-instruct": { + "id": "qwen3-next-80b-a3b-instruct", + "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-04-24", - "last_updated": "2025-04-24", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image"], - "output": ["image"] + "input": ["text"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 131072, + "output": 32768 }, "cost": { - "input": 5, - "output": 40, - "cache_read": 1.25 + "input": 0.5, + "output": 2 } }, - "openai-gpt-5.5": { - "id": "openai-gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-30", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 128000 + "output": 65536 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, "tiers": [ { - "input": 10, - "output": 45, - "cache_read": 1, + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, "tier": { "type": "context", - "size": 272000 + "size": 256000 } } ], "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 } } }, - "nvidia-nemotron-3-super-120b": { - "id": "nvidia-nemotron-3-super-120b", - "name": "Nemotron-3-Super-120B", - "family": "nemotron", - "attachment": false, + "qwen3.6-35b-a3b": { + "id": "qwen3.6-35b-a3b", + "name": "Qwen3.6 35B-A3B", + "description": "Open multimodal Qwen MoE for local agents that need vision, audio, and code", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2026-02", - "release_date": "2026-03-11", - "last_updated": "2026-04-16", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 32768 + "context": 262144, + "output": 65536 }, - "status": "beta", "cost": { - "input": 0.3, - "output": 0.65 + "input": 0.248, + "output": 1.485 } }, - "openai-gpt-5.4-pro": { - "id": "openai-gpt-5.4-pro", - "name": "GPT-5.4 pro", - "family": "gpt-pro", - "attachment": true, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "temperature": true, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 30, - "output": 180 + "input": 2.5, + "output": 7.5, + "cache_read": 0.5, + "cache_write": 3.125 } }, - "all-mini-lm-l6-v2": { - "id": "all-mini-lm-l6-v2", - "name": "All-MiniLM-L6-v2", - "family": "text-embedding", + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2021-08-30", - "last_updated": "2026-04-16", + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256, - "output": 384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.009, - "output": 0 + "input": 1.2, + "output": 6 } }, - "bge-m3": { - "id": "bge-m3", - "name": "BGE M3", - "family": "bge", + "qwen2-5-omni-7b": { + "id": "qwen2-5-omni-7b", + "name": "Qwen2.5-Omni 7B", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-01-30", - "last_updated": "2026-04-30", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-12", + "last_updated": "2024-12", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 1024 + "context": 32768, + "output": 2048 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.1, + "output": 0.4, + "input_audio": 6.76 } }, - "openai-gpt-5.1-codex-max": { - "id": "openai-gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", - "family": "gpt-codex", - "attachment": true, + "qwen3-8b": { + "id": "qwen3-8b", + "name": "Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.18, + "output": 0.7, + "reasoning": 2.1 } }, - "anthropic-claude-opus-4.6": { - "id": "anthropic-claude-opus-4.6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, - "reasoning": true, + "qwen2-5-14b-instruct": { + "id": "qwen2-5-14b-instruct", + "name": "Qwen2.5 14B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 0.5, - "cache_write": 6.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 0.5, - "cache_write": 6.25 - } + "input": 0.35, + "output": 1.4 } }, - "openai-gpt-4o": { - "id": "openai-gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, - "reasoning": false, + "qwen3-next-80b-a3b-thinking": { + "id": "qwen3-next-80b-a3b-thinking", + "name": "Qwen3-Next 80B-A3B (Thinking)", + "description": "Efficient Qwen thinking model for local reasoning, math, and coding agents", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.5, + "output": 6 } }, - "openai-gpt-5.4-mini": { - "id": "openai-gpt-5.4-mini", - "name": "GPT-5.4 mini", - "family": "gpt-mini", - "attachment": true, + "qvq-max": { + "id": "qvq-max", + "name": "QVQ Max", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qvq", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 1.2, + "output": 4.8 } }, - "openai-gpt-5": { - "id": "openai-gpt-5", - "name": "GPT-5", - "family": "gpt", + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "release_date": "2026-04-27", + "last_updated": "2026-04-27", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.1875, + "output": 1.125, + "cache_write": 0.234375 } }, - "arcee-trinity-large-thinking": { - "id": "arcee-trinity-large-thinking", - "name": "Trinity Large Thinking", - "family": "trinity", + "qwen2-5-vl-72b-instruct": { + "id": "qwen2-5-vl-72b-instruct", + "name": "Qwen2.5-VL 72B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-16", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 128000 + "context": 131072, + "output": 8192 }, - "status": "beta", "cost": { - "input": 0.25, - "output": 0.9, - "cache_read": 0.06 + "input": 2.8, + "output": 8.4 } }, - "mistral-nemo-instruct-2407": { - "id": "mistral-nemo-instruct-2407", - "name": "Mistral Nemo Instruct", - "family": "mistral", + "qwen3-vl-plus": { + "id": "qwen3-vl-plus", + "name": "Qwen3-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2025-04", + "release_date": "2025-09-23", + "last_updated": "2025-09-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.2, + "output": 1.6, + "reasoning": 4.8 } }, - "deepseek-v3": { - "id": "deepseek-v3", - "name": "DeepSeek V3", - "family": "deepseek", + "qwen-vl-ocr": { + "id": "qwen-vl-ocr", + "name": "Qwen-VL OCR", + "description": "OCR model for extracting structured text from documents and screenshots", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-12-26", - "last_updated": "2025-03-24", + "knowledge": "2024-04", + "release_date": "2024-10-28", + "last_updated": "2025-04-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 131072 + "context": 34096, + "output": 4096 + }, + "cost": { + "input": 0.72, + "output": 0.72 } }, - "bge-reranker-v2-m3": { - "id": "bge-reranker-v2-m3", - "name": "BGE Reranker v2 M3", - "family": "bge", + "qwen-mt-turbo": { + "id": "qwen-mt-turbo", + "name": "Qwen-MT Turbo", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-03-12", - "last_updated": "2026-04-30", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 8192, - "output": 1 - }, - "cost": { - "input": 0.01, - "output": 0 - } - }, - "anthropic-claude-3.7-sonnet": { - "id": "anthropic-claude-3.7-sonnet", - "name": "Claude 3.7 Sonnet", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "knowledge": "2024-04", + "release_date": "2025-01", + "last_updated": "2025-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 16384, + "output": 8192 }, - "status": "deprecated", "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.16, + "output": 0.49 } }, - "qwen3-tts-voicedesign": { - "id": "qwen3-tts-voicedesign", - "name": "Qwen3 TTS VoiceDesign", + "qwen-mt-plus": { + "id": "qwen-mt-plus", + "name": "Qwen-MT Plus", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2026-04-21", - "last_updated": "2026-04-30", + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-01", + "last_updated": "2025-01", "modalities": { "input": ["text"], - "output": ["audio"] - }, - "open_weights": true, - "limit": { - "context": 32768, - "output": 1 - } - }, - "openai-gpt-image-1.5": { - "id": "openai-gpt-image-1.5", - "name": "GPT Image 1.5", - "family": "gpt-image", - "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-11-25", - "last_updated": "2025-11-25", - "modalities": { - "input": ["text", "image"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 16384, + "output": 8192 }, "cost": { - "input": 5, - "output": 10, - "cache_read": 1 + "input": 2.46, + "output": 7.37 } }, - "openai-gpt-5.3-codex": { - "id": "openai-gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", - "attachment": true, + "qwen3.5-plus": { + "id": "qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.4, + "output": 2.4, + "reasoning": 2.4 } }, - "anthropic-claude-3.5-sonnet": { - "id": "anthropic-claude-3.5-sonnet", - "name": "Claude 3.5 Sonnet", - "family": "claude-sonnet", - "attachment": true, + "qwen-omni-turbo": { + "id": "qwen-omni-turbo", + "name": "Qwen-Omni Turbo", + "description": "Qwen omni model for text, vision, audio, and multimodal agent tasks", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2024-04", - "release_date": "2024-06-20", - "last_updated": "2024-10-22", + "release_date": "2025-01-19", + "last_updated": "2025-03-26", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 32768, + "output": 2048 }, - "status": "deprecated", "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.07, + "output": 0.27, + "input_audio": 4.44, + "output_audio": 8.89 } }, - "fal-ai/fast-sdxl": { - "id": "fal-ai/fast-sdxl", - "name": "Fast SDXL", - "family": "stable-diffusion", + "qwen2-5-72b-instruct": { + "id": "qwen2-5-72b-instruct", + "name": "Qwen2.5 72B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-07-26", - "last_updated": "2026-04-16", + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 131072, + "output": 8192 + }, + "cost": { + "input": 1.4, + "output": 5.6 } }, - "fal-ai/flux/schnell": { - "id": "fal-ai/flux/schnell", - "name": "FLUX.1 [schnell]", - "family": "flux", + "qwen-flash": { + "id": "qwen-flash", + "name": "Qwen Flash", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-08-01", - "last_updated": "2026-04-16", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "output": 32768 + }, + "cost": { + "input": 0.05, + "output": 0.4 } }, - "fal-ai/elevenlabs/tts/multilingual-v2": { - "id": "fal-ai/elevenlabs/tts/multilingual-v2", - "name": "ElevenLabs Multilingual TTS v2", - "family": "elevenlabs", + "qwen3-vl-235b-a22b": { + "id": "qwen3-vl-235b-a22b", + "name": "Qwen3-VL 235B-A22B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-08-22", - "last_updated": "2026-04-16", + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.7, + "output": 2.8, + "reasoning": 8.4 } }, - "fal-ai/stable-audio-25/text-to-audio": { - "id": "fal-ai/stable-audio-25/text-to-audio", - "name": "Stable Audio 2.5 (Text-to-Audio)", + "qwen3-vl-30b-a3b": { + "id": "qwen3-vl-30b-a3b", + "name": "Qwen3-VL 30B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-10-08", - "last_updated": "2026-04-16", + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens" + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, - "output": 0 + "context": 131072, + "output": 32768 + }, + "cost": { + "input": 0.2, + "output": 0.8, + "reasoning": 2.4 } - } - } - }, - "vultr": { - "id": "vultr", - "env": ["VULTR_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.vultrinference.com/v1", - "name": "Vultr", - "doc": "https://api.vultrinference.com/", - "models": { - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax M2.5", - "family": "minimax", + }, + "qwen-vl-max": { + "id": "qwen-vl-max", + "name": "Qwen-VL Max", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-02-11", - "last_updated": "2025-02-11", + "knowledge": "2024-04", + "release_date": "2024-04-08", + "last_updated": "2025-08-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 194000, - "output": 4096 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.8, + "output": 3.2 } }, - "GLM-5-FP8": { - "id": "GLM-5-FP8", - "name": "GLM 5 FP8", - "family": "glm", - "attachment": false, - "reasoning": false, + "qwen3.5-27b": { + "id": "qwen3.5-27b", + "name": "Qwen3.5 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.85, - "output": 3.1 + "input": 0.3, + "output": 2.4 } }, - "DeepSeek-V3.2": { - "id": "DeepSeek-V3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "qwen-max": { + "id": "qwen-max", + "name": "Qwen Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-04", + "release_date": "2024-04-03", + "last_updated": "2025-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 127000, - "output": 4096 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.55, - "output": 1.65 + "input": 1.6, + "output": 6.4 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "qwen3-235b-a22b": { + "id": "qwen3-235b-a22b", + "name": "Qwen3 235B-A22B", + "description": "Large open Qwen MoE for multilingual reasoning, coding, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 129000, - "output": 4096 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.7, + "output": 2.8, + "reasoning": 8.4 } }, - "Kimi-K2.5": { - "id": "Kimi-K2.5", - "name": "Kimi K2 Instruct", - "family": "kimi", + "qwen3-livetranslate-flash-realtime": { + "id": "qwen3-livetranslate-flash-realtime", + "name": "Qwen3-LiveTranslate Flash Realtime", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, "knowledge": "2024-04", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio", "video"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 254000, - "output": 32768 + "context": 53248, + "output": 4096 }, "cost": { - "input": 0.55, - "output": 2.75 + "input": 10, + "output": 10, + "input_audio": 10, + "output_audio": 38 } - } - } - }, - "alibaba-coding-plan-cn": { - "id": "alibaba-coding-plan-cn", - "env": ["ALIBABA_CODING_PLAN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://coding.dashscope.aliyuncs.com/v1", - "name": "Alibaba Coding Plan (China)", - "doc": "https://help.aliyun.com/zh/model-studio/coding-plan", - "models": { - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3 Coder Plus", + }, + "qwen3.6-27b": { + "id": "qwen3.6-27b", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, - "reasoning": false, + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, + "context": 262144, "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.6, + "output": 3.6 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "qwen3.5-35b-a3b": { + "id": "qwen3.5-35b-a3b", + "name": "Qwen3.5 35B-A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 32768 + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.25, + "output": 2 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "qwen3-coder-480b-a35b-instruct": { + "id": "qwen3-coder-480b-a35b-instruct", + "name": "Qwen3-Coder 480B-A35B Instruct", + "description": "Open Qwen coding heavyweight for repository reasoning and agentic engineering", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 16384 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 1.5, + "output": 7.5 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", + "qwq-plus": { + "id": "qwq-plus", + "name": "QwQ Plus", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2024-04", + "release_date": "2025-03-05", + "last_updated": "2025-03-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.8, + "output": 2.4 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "qwen2-5-32b-instruct": { + "id": "qwen2-5-32b-instruct", + "name": "Qwen2.5 32B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 24576 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.7, + "output": 2.8 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, + "context": 262144, "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.6, + "output": 3.6 } }, - "qwen3-max-2026-01-23": { - "id": "qwen3-max-2026-01-23", - "name": "Qwen3 Max", + "qwen3-coder-flash": { + "id": "qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, "knowledge": "2025-04", - "release_date": "2026-01-23", - "last_updated": "2026-01-23", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.3, + "output": 1.5 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "Qwen3 Coder Next", + "qwen3-14b": { + "id": "qwen3-14b", + "name": "Qwen3 14B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2026-02-03", - "last_updated": "2026-02-03", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.35, + "output": 1.4, + "reasoning": 4.2 } }, - "qwen3.5-plus": { - "id": "qwen3.5-plus", - "name": "Qwen3.5 Plus", + "qwen3-asr-flash": { + "id": "qwen3-asr-flash", + "name": "Qwen3-ASR Flash", + "description": "Speech transcription model for accurate audio-to-text and captioning workflows", "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-09-08", + "last_updated": "2025-09-08", "modalities": { - "input": ["text", "image"], + "input": ["audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 53248, + "output": 4096 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.035, + "output": 0.035 } - } - } - }, - "mistral": { - "id": "mistral", - "env": ["MISTRAL_API_KEY"], - "npm": "@ai-sdk/mistral", - "name": "Mistral", - "doc": "https://docs.mistral.ai/getting-started/models/", - "models": { - "mistral-small-latest": { - "id": "mistral-small-latest", - "name": "Mistral Small (latest)", - "family": "mistral-small", - "attachment": true, + }, + "qwen-turbo": { + "id": "qwen-turbo", + "name": "Qwen Turbo", + "description": "Efficient Qwen model for fast chat, extraction, and high-volume workloads", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2024-04", + "release_date": "2024-11-01", + "last_updated": "2025-04-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.05, + "output": 0.2, + "reasoning": 0.5 } }, - "mistral-nemo": { - "id": "mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", + "qwen2-5-7b-instruct": { + "id": "qwen2-5-7b-instruct", + "name": "Qwen2.5 7B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.175, + "output": 0.7 } }, - "mistral-large-2512": { - "id": "mistral-large-2512", - "name": "Mistral Large 3", - "family": "mistral-large", - "attachment": true, + "qwen2-5-vl-7b-instruct": { + "id": "qwen2-5-vl-7b-instruct", + "name": "Qwen2.5-VL 7B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2025-12-02", + "knowledge": "2024-04", + "release_date": "2024-09", + "last_updated": "2024-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 131072, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.35, + "output": 1.05 } }, - "labs-devstral-small-2512": { - "id": "labs-devstral-small-2512", - "name": "Devstral Small 2", - "family": "devstral", + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "knowledge": "2025-04", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.3, + "output": 7.8, + "cache_read": 0.13, + "cache_write": 1.625 } }, - "devstral-2512": { - "id": "devstral-2512", - "name": "Devstral 2", - "family": "devstral", - "attachment": false, - "reasoning": false, + "qwen3.5-122b-a10b": { + "id": "qwen3.5-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 65536 }, "cost": { "input": 0.4, - "output": 2 + "output": 3.2 } }, - "magistral-medium-latest": { - "id": "magistral-medium-latest", - "name": "Magistral Medium (latest)", - "family": "magistral-medium", + "qwen-vl-plus": { + "id": "qwen-vl-plus", + "name": "Qwen-VL Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-03-17", - "last_updated": "2025-03-20", + "knowledge": "2024-04", + "release_date": "2024-01-25", + "last_updated": "2025-08-15", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 131072, + "output": 8192 }, "cost": { - "input": 2, - "output": 5 + "input": 0.21, + "output": 0.63 } }, - "open-mixtral-8x7b": { - "id": "open-mixtral-8x7b", - "name": "Mixtral 8x7B", - "family": "mixtral", + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Earlier Qwen multimodal workhorse for million-token agent and document tasks", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-01", - "release_date": "2023-12-11", - "last_updated": "2023-12-11", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 32000 + "context": 1000000, + "output": 65536 }, "cost": { - "input": 0.7, - "output": 0.7 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "cache_write": 0.625, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.2, + "cache_write": 2.5 + } + } + } + } + }, + "databricks": { + "id": "databricks", + "env": ["DATABRICKS_HOST", "DATABRICKS_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://${DATABRICKS_HOST}/ai-gateway/mlflow/v1", + "name": "Databricks", + "doc": "https://docs.databricks.com/aws/en/machine-learning/foundation-models/", + "models": { + "databricks-claude-opus-4-7": { + "id": "databricks-claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "pixtral-large-latest": { - "id": "pixtral-large-latest", - "name": "Pixtral Large (latest)", - "family": "pixtral", + "databricks-gpt-5-4": { + "id": "databricks-gpt-5-4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2024-11-04", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 1050000, + "input": 922000, "output": 128000 }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, "cost": { - "input": 2, - "output": 6 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "mistral-large-2411": { - "id": "mistral-large-2411", - "name": "Mistral Large 2.1", - "family": "mistral-large", - "attachment": false, - "reasoning": false, + "databricks-gemini-3-flash": { + "id": "databricks-gemini-3-flash", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2024-11-04", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 2, - "output": 6 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 } }, - "codestral-latest": { - "id": "codestral-latest", - "name": "Codestral (latest)", - "family": "codestral", - "attachment": false, - "reasoning": false, + "databricks-claude-opus-4-5": { + "id": "databricks-claude-opus-4-5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-05-29", - "last_updated": "2025-01-04", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 4096 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "mistral-large-latest": { - "id": "mistral-large-latest", - "name": "Mistral Large (latest)", - "family": "mistral-large", + "databricks-gpt-5-nano": { + "id": "databricks-gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2025-12-02", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "mistral-small-2506": { - "id": "mistral-small-2506", - "name": "Mistral Small 3.2", - "family": "mistral-small", - "attachment": false, - "reasoning": false, + "databricks-gpt-5-mini": { + "id": "databricks-gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-06-20", - "last_updated": "2025-06-20", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "pixtral-12b": { - "id": "pixtral-12b", - "name": "Pixtral 12B", - "family": "pixtral", + "databricks-gpt-5": { + "id": "databricks-gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-09-01", - "last_updated": "2024-09-01", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "ministral-8b-latest": { - "id": "ministral-8b-latest", - "name": "Ministral 8B (latest)", - "family": "ministral", - "attachment": false, - "reasoning": false, + "databricks-gemini-2-5-pro": { + "id": "databricks-gemini-2-5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-04", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "mistral-embed": { - "id": "mistral-embed", - "name": "Mistral Embed", - "family": "mistral-embed", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-12-11", - "last_updated": "2023-12-11", + "databricks-gemini-3-1-pro": { + "id": "databricks-gemini-3-1-pro", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "output": 3072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "magistral-small": { - "id": "magistral-small", - "name": "Magistral Small", - "family": "magistral-small", - "attachment": false, + "databricks-gemini-2-5-flash": { + "id": "databricks-gemini-2-5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-03-17", - "last_updated": "2025-03-17", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 } }, - "mistral-small-2603": { - "id": "mistral-small-2603", - "name": "Mistral Small 4", - "family": "mistral-small", + "databricks-claude-sonnet-4": { + "id": "databricks-claude-sonnet-4", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "ministral-3b-latest": { - "id": "ministral-3b-latest", - "name": "Ministral 3B (latest)", - "family": "ministral", + "databricks-glm-5-2": { + "id": "databricks-glm-5-2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-04", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 1000000, + "output": 131072 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 1.4, + "output": 4.4, + "cache_read": 0.26 } }, - "open-mixtral-8x22b": { - "id": "open-mixtral-8x22b", - "name": "Mixtral 8x22B", - "family": "mixtral", - "attachment": false, - "reasoning": false, + "databricks-claude-haiku-4-5": { + "id": "databricks-claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-17", - "last_updated": "2024-04-17", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, + "context": 200000, "output": 64000 }, "cost": { - "input": 2, - "output": 6 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "mistral-medium-2604": { - "id": "mistral-medium-2604", - "name": "Mistral Medium 3.5", - "family": "mistral-medium", + "databricks-gpt-5-4-nano": { + "id": "databricks-gpt-5-4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-29", - "last_updated": "2026-04-29", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1.5, - "output": 7.5 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "devstral-small-2505": { - "id": "devstral-small-2505", - "name": "Devstral Small 2505", - "family": "devstral", - "attachment": false, - "reasoning": false, + "databricks-claude-opus-4-6": { + "id": "databricks-claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 1000000, "output": 128000 }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "devstral-medium-2507": { - "id": "devstral-medium-2507", - "name": "Devstral Medium", - "family": "devstral", - "attachment": false, - "reasoning": false, + "databricks-gpt-5-1": { + "id": "databricks-gpt-5-1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "open-mistral-7b": { - "id": "open-mistral-7b", - "name": "Mistral 7B", - "family": "mistral", - "attachment": false, - "reasoning": false, + "databricks-claude-sonnet-4-5": { + "id": "databricks-claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-09-27", - "last_updated": "2023-09-27", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8000, - "output": 8000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.25, - "output": 0.25 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "devstral-medium-latest": { - "id": "devstral-medium-latest", - "name": "Devstral 2 (latest)", - "family": "devstral", - "attachment": false, - "reasoning": false, + "databricks-claude-opus-4-1": { + "id": "databricks-claude-opus-4-1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.4, - "output": 2 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "mistral-medium-2505": { - "id": "mistral-medium-2505", - "name": "Mistral Medium 3", - "family": "mistral-medium", + "databricks-gpt-5-5": { + "id": "databricks-gpt-5-5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } }, "cost": { - "input": 0.4, - "output": 2 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } }, - "devstral-small-2507": { - "id": "devstral-small-2507", - "name": "Devstral Small", - "family": "devstral", - "attachment": false, - "reasoning": false, + "databricks-gemini-3-1-flash-lite": { + "id": "databricks-gemini-3-1-flash-lite", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 } }, - "mistral-medium-2508": { - "id": "mistral-medium-2508", - "name": "Mistral Medium 3.1", - "family": "mistral-medium", + "databricks-gemini-3-pro": { + "id": "databricks-gemini-3-pro", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-08-12", - "last_updated": "2025-08-12", + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.4, - "output": 2 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "mistral-medium-latest": { - "id": "mistral-medium-latest", - "name": "Mistral Medium (latest)", - "family": "mistral-medium", + "databricks-kimi-k2-7-code": { + "id": "databricks-kimi-k2-7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-29", - "last_updated": "2026-04-29", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, @@ -88738,305 +139378,461 @@ "output": 262144 }, "cost": { - "input": 1.5, - "output": 7.5 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } - } - } - }, - "ovhcloud": { - "id": "ovhcloud", - "env": ["OVHCLOUD_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://oai.endpoints.kepler.ai.cloud.ovh.net/v1", - "name": "OVHcloud AI Endpoints", - "doc": "https://www.ovhcloud.com/en/public-cloud/ai-endpoints/catalog//", - "models": { - "meta-llama-3_3-70b-instruct": { - "id": "meta-llama-3_3-70b-instruct", - "name": "Meta-Llama-3_3-70B-Instruct", - "attachment": false, - "reasoning": false, + }, + "databricks-claude-sonnet-4-6": { + "id": "databricks-claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-01", - "last_updated": "2025-04-01", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.74, - "output": 0.74 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "mistral-7b-instruct-v0.3": { - "id": "mistral-7b-instruct-v0.3", - "name": "Mistral-7B-Instruct-v0.3", + "databricks-gpt-oss-20b": { + "id": "databricks-gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-04-01", - "last_updated": "2025-04-01", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.11, - "output": 0.11 + "input": 0.05, + "output": 0.2 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3-32B", + "databricks-gpt-oss-120b": { + "id": "databricks-gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-07-16", - "last_updated": "2025-07-16", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, + "context": 131072, "output": 32768 }, "cost": { - "input": 0.09, - "output": 0.25 + "input": 0.072, + "output": 0.28 } }, - "qwen2.5-vl-72b-instruct": { - "id": "qwen2.5-vl-72b-instruct", - "name": "Qwen2.5-VL-72B-Instruct", + "databricks-gpt-5-2": { + "id": "databricks-gpt-5-2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-03-31", - "last_updated": "2025-03-31", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 1.01, - "output": 1.01 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3-Coder-30B-A3B-Instruct", + "databricks-gpt-5-4-mini": { + "id": "databricks-gpt-5-4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, + "cost": { + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 + } + } + } + }, + "crof": { + "id": "crof", + "env": ["CROF_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://crof.ai/v1", + "name": "CrofAI", + "doc": "https://crof.ai/docs", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-10-28", - "last_updated": "2025-10-28", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.07, - "output": 0.26 + "input": 0.12, + "output": 0.21, + "cache_read": 0.003 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "gpt-oss-20b", + "minimax-m2.5": { + "id": "minimax-m2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 204800, "output": 131072 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 0.05, - "output": 0.18 + "input": 0.11, + "output": 0.95, + "cache_read": 0.02, + "cache_write": 0.375 } }, - "mistral-small-3.2-24b-instruct-2506": { - "id": "mistral-small-3.2-24b-instruct-2506", - "name": "Mistral-Small-3.2-24B-Instruct-2506", - "attachment": true, + "greg-2-ultra": { + "id": "greg-2-ultra", + "name": "Greg 2 Ultra", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-07-16", - "last_updated": "2025-07-16", + "release_date": "2026-06-14", + "last_updated": "2026-06-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 229376, + "output": 229376 }, "cost": { - "input": 0.1, - "output": 0.31 + "input": 3, + "output": 10, + "cache_read": 0.5 } }, - "qwen3.5-9b": { - "id": "qwen3.5-9b", - "name": "Qwen3.5-9B", - "attachment": true, - "reasoning": false, + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Mature GLM model for dependable coding, reasoning, and structured agent tasks", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 202752, + "output": 202752 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.1, - "output": 0.15 + "input": 0.25, + "output": 1.1, + "cache_read": 0.05, + "cache_write": 0 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "gpt-oss-120b", + "deepseek-v4-pro-lightning": { + "id": "deepseek-v4-pro-lightning", + "name": "DeepSeek V4 Pro Lightning", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 1000000, "output": 131072 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 0.09, - "output": 0.47 + "input": 0.8, + "output": 1.6, + "cache_read": 0.02 } }, - "mistral-nemo-instruct-2407": { - "id": "mistral-nemo-instruct-2407", - "name": "Mistral-Nemo-Instruct-2407", + "greg-rp": { + "id": "greg-rp", + "name": "Greg (Roleplay)", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2024-11-20", - "last_updated": "2024-11-20", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 229376, + "output": 229376 }, "cost": { - "input": 0.14, - "output": 0.14 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } }, - "llama-3.1-8b-instruct": { - "id": "llama-3.1-8b-instruct", - "name": "Llama-3.1-8B-Instruct", - "attachment": false, - "reasoning": false, + "gemma-4-31b-it": { + "id": "gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-06-11", - "last_updated": "2025-06-11", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.11, - "output": 0.11 + "input": 0.1, + "output": 0.3, + "cache_read": 0.02 } - } - } - }, - "friendli": { - "id": "friendli", - "env": ["FRIENDLI_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.friendli.ai/serverless/v1", - "name": "Friendli", - "doc": "https://friendli.ai/docs/guides/serverless_endpoints/introduction", - "models": { - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + }, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "release_date": "2025-07-29", - "last_updated": "2026-01-29", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, @@ -89044,17 +139840,28 @@ "context": 262144, "output": 262144 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 0.2, - "output": 0.8 + "input": 0.55, + "output": 2.25, + "cache_read": 0.05 } }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", + "glm-5.1": { + "id": "glm-5.1", "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -89072,326 +139879,446 @@ "context": 202752, "output": 202752 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.45, + "output": 2.15, + "cache_read": 0.08, + "cache_write": 0 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "GLM-5", - "family": "glm", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 202752 + "context": 1000000, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.5 + "input": 0.35, + "output": 0.8, + "cache_read": 0.003 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2024-08-01", - "last_updated": "2025-12-23", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 1000000, "output": 131072 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 0.6, - "output": 0.6 + "input": 0.5, + "output": 2.2, + "cache_read": 0.08 } }, - "meta-llama/Llama-3.1-8B-Instruct": { - "id": "meta-llama/Llama-3.1-8B-Instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "greg-2-super": { + "id": "greg-2-super", + "name": "Greg 2 Super", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2024-08-01", - "last_updated": "2025-12-23", + "release_date": "2026-06-14", + "last_updated": "2026-06-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8000 + "context": 229376, + "output": 229376 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 1.5, + "output": 5, + "cache_read": 0.25 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "kimi-k2.5-lightning": { + "id": "kimi-k2.5-lightning", + "name": "Kimi K2.5 (Lightning)", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "temperature": false, + "release_date": "2026-02-06", + "last_updated": "2026-02-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 196608, - "output": 196608 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 1, + "output": 3, + "cache_read": 0.2 } - } - } - }, - "cortecs": { - "id": "cortecs", - "env": ["CORTECS_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.cortecs.ai/v1", - "name": "Cortecs", - "doc": "https://api.cortecs.ai/v1/models", - "models": { - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "MiniMax-m2.7", - "family": "minimax", + }, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 196072 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.47, - "output": 1.4 + "input": 0.35, + "output": 1.7, + "cache_read": 0.07 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 1.09, - "output": 5.43 + "input": 0.5, + "output": 1.99, + "cache_read": 0.05 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", + "qwen3.6-27b": { + "id": "qwen3.6-27b", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.062, - "output": 0.408 + "input": 0.2, + "output": 1.5, + "cache_read": 0.04 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi-thinking", - "attachment": false, + "qwen3.5-9b": { + "id": "qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-13", + "last_updated": "2026-03-13", + "modalities": { + "input": ["text", "image", "video", "audio"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.04, + "output": 0.15, + "cache_read": 0.008 + } + }, + "qwen3.5-397b-a17b": { + "id": "qwen3.5-397b-a17b", + "name": "Qwen3.5 397B-A17B", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.55, - "output": 2.76 + "input": 0.35, + "output": 1.75, + "cache_read": 0.07 } }, - "deepseek-v3-0324": { - "id": "deepseek-v3-0324", - "name": "DeepSeek V3 0324", - "family": "deepseek", + "glm-4.7-flash": { + "id": "glm-4.7-flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 128000 + "context": 202752, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 0.551, - "output": 1.654 + "input": 0.04, + "output": 0.3, + "cache_read": 0.008, + "cache_write": 0 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM 4.7", - "family": "glm", + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 198000, - "output": 198000 - }, - "cost": { - "input": 0.45, - "output": 2.23 - } - }, - "claude-opus4-7": { - "id": "claude-opus4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "context": 1048576, + "output": 131072 }, - "open_weights": false, - "limit": { - "context": 1000000, - "output": 128000 + "provider": { + "npm": "@ai-sdk/openai-compatible" }, "cost": { - "input": 5.6, - "output": 27.99, - "cache_read": 0.56, - "cache_write": 6.99 + "input": 0.4, + "output": 0.8, + "cache_read": 0.003, + "tiers": [ + { + "input": 2, + "output": 6, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 6, + "cache_read": 0.4 + } } }, "glm-5": { "id": "glm-5", - "name": "GLM 5", + "name": "GLM-5", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] @@ -89401,988 +140328,1315 @@ "context": 202752, "output": 202752 }, + "provider": { + "npm": "@ai-sdk/openai-compatible" + }, "cost": { - "input": 1.08, - "output": 3.44 + "input": 0.48, + "output": 1.9, + "cache_read": 0.1, + "cache_write": 0 } }, - "nova-pro-v1": { - "id": "nova-pro-v1", - "name": "Nova Pro 1.0", - "family": "nova-pro", + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 5000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 1.016, - "output": 4.061 + "input": 0.18, + "output": 0.35, + "cache_read": 0.04 } }, - "devstral-2512": { - "id": "devstral-2512", - "name": "Devstral 2 2512", + "greg-1-mini": { + "id": "greg-1-mini", + "name": "Greg 1 Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "release_date": "2026-01-27", + "last_updated": "2026-01-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 229376, + "output": 229376 }, "cost": { - "input": 0, - "output": 0 + "input": 0.07, + "output": 0.15, + "cache_read": 0.01 } - }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", - "family": "qwen", - "attachment": false, + } + } + }, + "fastrouter": { + "id": "fastrouter", + "env": ["FASTROUTER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://go.fastrouter.ai/api/v1", + "name": "FastRouter", + "doc": "https://fastrouter.ai/models", + "models": { + "wanx/wan-v2-6": { + "id": "wanx/wan-v2-6", + "name": "Wan 2.6", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "tool_call": false, + "temperature": false, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 - }, - "cost": { - "input": 0.099, - "output": 0.33 + "context": 400000, + "output": 0 } }, - "codestral-2508": { - "id": "codestral-2508", - "name": "Codestral 2508", - "family": "mistral", + "moonshotai/kimi-k2": { + "id": "moonshotai/kimi-k2", + "name": "Kimi K2", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "knowledge": "2024-10", + "release_date": "2025-07-11", + "last_updated": "2025-07-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 0.9, - "cache_read": 0.03 + "input": 0.55, + "output": 2.2 } }, - "claude-4-5-sonnet": { - "id": "claude-4-5-sonnet", - "name": "Claude 4.5 Sonnet", - "family": "claude-sonnet", + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 200000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 3.259, - "output": 16.296 + "input": 0.75, + "output": 3.5 } }, - "kimi-k2-instruct": { - "id": "kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", + "google/imagen-4.0-fast": { + "id": "google/imagen-4.0-fast", + "name": "Imagen 4 Fast", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-07-11", - "last_updated": "2025-09-05", + "tool_call": false, + "temperature": false, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131000, - "output": 131000 - }, - "cost": { - "input": 0.551, - "output": 2.646 + "context": 480, + "output": 0 } }, - "nemotron-3-super-120b-a12b": { - "id": "nemotron-3-super-120b-a12b", - "name": "Nemotron 3 Super 120B A12B", - "attachment": false, + "google/gemini-2.5-pro": { + "id": "google/gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.266, - "output": 0.799 + "input": 1.25, + "output": 10, + "cache_read": 0.31 } }, - "minimax-m2": { - "id": "minimax-m2", - "name": "MiniMax-M2", - "family": "minimax", - "attachment": false, + "google/gemini-2.5-flash": { + "id": "google/gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 400000, - "output": 400000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.39, - "output": 1.57 + "input": 0.3, + "output": 2.5, + "cache_read": 0.0375 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": false, - "reasoning": false, + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-17", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 65535 + "output": 65536 }, "cost": { - "input": 1.654, - "output": 11.024 + "input": 1.5, + "output": 9 } }, - "claude-opus4-6": { - "id": "claude-opus4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "google/veo3.1-lite": { + "id": "google/veo3.1-lite", + "name": "Veo 3.1 Lite", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 1000000 - }, - "cost": { - "input": 5.98, - "output": 29.89 + "context": 400000, + "output": 0 } }, - "devstral-small-2512": { - "id": "devstral-small-2512", - "name": "Devstral Small 2 2512", - "attachment": false, - "reasoning": false, + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0, + "input": 0.13, + "output": 0.38 + } + }, + "google/veo3.1": { + "id": "google/veo3.1", + "name": "Veo 3.1", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", + "modalities": { + "input": ["text", "image"], + "output": ["video"] + }, + "open_weights": false, + "limit": { + "context": 400000, "output": 0 } }, - "minimax-m2.1": { - "id": "minimax-m2.1", - "name": "MiniMax-M2.1", - "family": "minimax", + "google/imagen-4.0-ultra": { + "id": "google/imagen-4.0-ultra", + "name": "Imagen 4 Ultra", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "imagen", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", + "modalities": { + "input": ["text"], + "output": ["image"] }, + "open_weights": false, + "limit": { + "context": 480, + "output": 0 + } + }, + "google/gemini-3-pro-image-preview": { + "id": "google/gemini-3-pro-image-preview", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2025-01", + "release_date": "2025-11-20", + "last_updated": "2025-11-20", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196000, - "output": 196000 + "context": 65536, + "output": 32768 }, "cost": { - "input": 0.34, - "output": 1.34 + "input": 2, + "output": 12 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, + "google/gemini-3.1-flash-image-preview": { + "id": "google/gemini-3.1-flash-image-preview", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, "temperature": true, - "release_date": "2026-04-14", - "last_updated": "2026-04-14", + "knowledge": "2025-01", + "release_date": "2026-02-26", + "last_updated": "2026-02-26", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 65536, + "output": 65536 }, "cost": { - "input": 1.31, - "output": 4.1, - "cache_read": 0.24 + "input": 0.5, + "output": 3 } }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM 4.5", - "family": "glm", - "attachment": false, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-29", - "last_updated": "2025-07-29", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.67, - "output": 2.46 + "input": 2, + "output": 12 } }, - "claude-opus4-5": { - "id": "claude-opus4-5", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "google/veo3.1-fast": { + "id": "google/veo3.1-fast", + "name": "Veo 3.1 Fast", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "veo", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", + "modalities": { + "input": ["text", "image"], + "output": ["video"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "output": 0 + } + }, + "x-ai/grok-4.3": { + "id": "x-ai/grok-4.3", + "name": "Grok 4.3", + "description": "xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk", + "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 200000 + "context": 1000000, + "output": 30000 }, "cost": { - "input": 5.98, - "output": 29.89 + "input": 1.25, + "output": 2.5 } }, - "claude-sonnet-4": { - "id": "claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "x-ai/grok-4": { + "id": "x-ai/grok-4", + "name": "Grok 4", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "family": "grok", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2025-07", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 256000, "output": 64000 }, "cost": { - "input": 3.307, - "output": 16.536 + "input": 3, + "output": 15, + "cache_read": 0.75, + "cache_write": 15 } }, - "qwen3-next-80b-a3b-thinking": { - "id": "qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", - "attachment": false, + "x-ai/grok-build-0.1": { + "id": "x-ai/grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Fast Grok coding model tuned for agentic engineering and iterative edits", + "family": "grok-build", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-11", - "last_updated": "2025-09-11", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.164, - "output": 1.311 + "input": 1, + "output": 2 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM 4.5 Air", - "family": "glm-air", + "z-ai/glm-5.1": { + "id": "z-ai/glm-5.1", + "name": "GLM-5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-01", - "last_updated": "2025-08-01", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, + "context": 200000, "output": 131072 }, "cost": { - "input": 0.22, - "output": 1.34 + "input": 1.05, + "output": 3.5 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-thinking", + "z-ai/glm-5": { + "id": "z-ai/glm-5", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.81, - "output": 3.54, - "cache_read": 0.2 + "input": 0.95, + "output": 3.15 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "Qwen3 Coder Next 80B", - "family": "qwen", - "attachment": false, + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-02-04", - "last_updated": "2026-02-04", + "knowledge": "2024-10-01", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 65536 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.158, - "output": 0.84 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "claude-4-6-sonnet": { - "id": "claude-4-6-sonnet", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, + "structured_output": true, + "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 272000, + "output": 128000 + }, + "cost": { + "input": 0.2, + "output": 1.25 + } + }, + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 1000000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 3.59, - "output": 17.92 + "input": 1.75, + "output": 14 } }, - "qwen3-coder-480b-a35b-instruct": { - "id": "qwen3-coder-480b-a35b-instruct", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", - "attachment": false, + "openai/gpt-realtime-1.5": { + "id": "openai/gpt-realtime-1.5", + "name": "GPT Realtime 1.5", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "audio", "image"], + "output": ["text", "audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.441, - "output": 1.984 + "input": 4, + "output": 16 } }, - "mixtral-8x7B-instruct-v0.1": { - "id": "mixtral-8x7B-instruct-v0.1", - "name": "Mixtral 8x7B Instruct v0.1", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2023-12-11", - "last_updated": "2023-12-11", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32000, - "output": 32000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.438, - "output": 0.68 + "input": 0.15, + "output": 0.6 } }, - "hermes-4-70b": { - "id": "hermes-4-70b", - "name": "Hermes 4 70B", - "attachment": false, + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 400000, + "input": 272000, "output": 128000 }, "cost": { - "input": 0.116, - "output": 0.358 + "input": 0.75, + "output": 4.5 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 196608 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.32, - "output": 1.18 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-10-01", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.266, - "output": 0.444 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "intellect-3": { - "id": "intellect-3", - "name": "INTELLECT 3", + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-26", - "last_updated": "2025-11-26", + "knowledge": "2024-10-01", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 400000, "output": 128000 }, "cost": { - "input": 0.219, - "output": 1.202 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "glm-4.7-flash": { - "id": "glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm", - "attachment": false, + "openai/gpt-5.5-pro": { + "id": "openai/gpt-5.5-pro", + "name": "GPT-5.5 Pro", + "description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding", + "family": "gpt-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-08", - "last_updated": "2025-08-08", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 203000, - "output": 203000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.09, - "output": 0.53 + "input": 30, + "output": 180 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT Oss 120b", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-01", "release_date": "2025-08-05", "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 65536 + }, + "cost": { + "input": 0.05, + "output": 0.2 + } + }, + "openai/gpt-image-2": { + "id": "openai/gpt-image-2", + "name": "GPT Image 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gpt-image", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["image"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "output": 0 + } + }, + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", + "modalities": { + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 30 } }, - "qwen-2.5-72b-instruct": { - "id": "qwen-2.5-72b-instruct", - "name": "Qwen2.5 72B Instruct", - "family": "qwen", - "attachment": false, + "bytedance/seedance-2": { + "id": "bytedance/seedance-2", + "name": "Seedance 2", + "description": "Video model for prompt-guided generation, editing, and motion workflows", + "family": "seed", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "tool_call": false, + "temperature": false, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["video"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 33000, - "output": 33000 - }, - "cost": { - "input": 0.062, - "output": 0.231 + "context": 4096, + "output": 0 } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek R1 0528", - "family": "deepseek-thinking", - "attachment": false, + "anthropic/claude-sonnet-4": { + "id": "anthropic/claude-sonnet-4", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.585, - "output": 2.307 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT 4.1", - "family": "gpt", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2.354, - "output": 9.417 + "input": 5, + "output": 25 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", + "anthropic/claude-opus-4.1": { + "id": "anthropic/claude-opus-4.1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.656, - "output": 2.731 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "llama-3.1-405b-instruct": { - "id": "llama-3.1-405b-instruct", - "name": "Llama 3.1 405B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15 } }, - "qwen3.5-122b-a10b": { - "id": "qwen3.5-122b-a10b", - "name": "Qwen3.5 122B A10B", - "family": "qwen", + "sarvam/sarvam-105b": { + "id": "sarvam/sarvam-105b", + "name": "Sarvam 105B", + "description": "Flagship Indian-language reasoning model for enterprise multilingual applications", + "family": "sarvam", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "release_date": "2025-09-01", + "last_updated": "2025-09-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.444, - "output": 3.106 + "input": 0.04, + "output": 0.16 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + "sarvam/sarvam-30b": { + "id": "sarvam/sarvam-30b", + "name": "Sarvam 30B", + "description": "Efficient Indian-language reasoning model for chat, coding, and multilingual work", + "family": "sarvam", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-02-18", + "last_updated": "2026-02-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.089, - "output": 0.275 + "input": 0.02, + "output": 0.1 } }, - "mistral-large-2512": { - "id": "mistral-large-2512", - "name": "Mistral Large 3 2512", - "family": "mistral-large", - "attachment": true, - "reasoning": false, - "tool_call": true, + "deepseek-ai/deepseek-r1-distill-llama-70b": { + "id": "deepseek-ai/deepseek-r1-distill-llama-70b", + "name": "DeepSeek R1 Distill Llama 70B", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-10", + "release_date": "2025-01-23", + "last_updated": "2025-01-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.5, - "output": 1.5, - "cache_read": 0.05 + "input": 0.03, + "output": 0.14 } }, - "qwen3.5-397b-a17b": { - "id": "qwen3.5-397b-a17b", - "name": "Qwen3.5 397B A17B", + "qwen/qwen3-coder": { + "id": "qwen/qwen3-coder", + "name": "Qwen3 Coder", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "knowledge": "2025-04", + "release_date": "2025-07-23", + "last_updated": "2025-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 250000, - "output": 250000 + "context": 262144, + "output": 66536 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.3, + "output": 1.2 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, "knowledge": "2025-05", "release_date": "2026-04-24", @@ -90393,3714 +141647,4803 @@ }, "open_weights": true, "limit": { - "context": 1048576, + "context": 1000000, "output": 384000 }, "cost": { - "input": 0.133, - "output": 0.266, - "cache_read": 0.028 + "input": 1.74, + "output": 3.48 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", + "minimax/minimax-m2.7-highspeed": { + "id": "minimax/minimax-m2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1048576, - "output": 384000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 1.553, - "output": 3.106, - "cache_read": 0.145 + "input": 0.6, + "output": 2.4 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3 Coder 30B A3B Instruct", - "family": "qwen", + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.053, - "output": 0.222 + "input": 0.3, + "output": 1.2 } - } - } - }, - "siliconflow": { - "id": "siliconflow", - "env": ["SILICONFLOW_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.siliconflow.com/v1", - "name": "SiliconFlow", - "doc": "https://cloud.siliconflow.com/models", - "models": { - "nex-agi/DeepSeek-V3.1-Nex-N1": { - "id": "nex-agi/DeepSeek-V3.1-Nex-N1", - "name": "nex-agi/DeepSeek-V3.1-Nex-N1", - "family": "deepseek", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-01-01", - "last_updated": "2025-11-25", + }, + "leonardo-ai/lucid-origin": { + "id": "leonardo-ai/lucid-origin", + "name": "Lucid Origin", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "lucid", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 - }, - "cost": { - "input": 0.5, - "output": 2 + "context": 4096, + "output": 0 } }, - "Qwen/Qwen2.5-VL-72B-Instruct": { - "id": "Qwen/Qwen2.5-VL-72B-Instruct", - "name": "Qwen/Qwen2.5-VL-72B-Instruct", - "family": "qwen", + "leonardo-ai/lucid-realism": { + "id": "leonardo-ai/lucid-realism", + "name": "Lucid Realism", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "lucid", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-01-28", - "last_updated": "2025-11-25", + "tool_call": false, + "temperature": false, + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["image"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 4000 - }, - "cost": { - "input": 0.59, - "output": 0.59 + "context": 4096, + "output": 0 } - }, - "Qwen/Qwen3-VL-32B-Thinking": { - "id": "Qwen/Qwen3-VL-32B-Thinking", - "name": "Qwen/Qwen3-VL-32B-Thinking", - "family": "qwen", + } + } + }, + "abliteration-ai": { + "id": "abliteration-ai", + "env": ["ABLIT_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.abliteration.ai/v1", + "name": "abliteration.ai", + "doc": "https://docs.abliteration.ai/models", + "models": { + "abliterated-model": { + "id": "abliterated-model", + "name": "Abliterated Model", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "release_date": "2025-10-21", - "last_updated": "2025-11-25", + "release_date": "2026-01-06", + "last_updated": "2026-01-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 150000, + "input": 150000, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 1.5 + "input": 3, + "output": 3 } - }, - "Qwen/Qwen3-30B-A3B-Thinking-2507": { - "id": "Qwen/Qwen3-30B-A3B-Thinking-2507", - "name": "Qwen/Qwen3-30B-A3B-Thinking-2507", - "family": "qwen", + } + } + }, + "xpersona": { + "id": "xpersona", + "env": ["XPERSONA_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://www.xpersona.co/v1", + "name": "Xpersona", + "doc": "https://www.xpersona.co/docs", + "models": { + "xpersona-gpt-5.5": { + "id": "xpersona-gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": true, - "release_date": "2025-07-31", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-12-30", + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 131000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.09, - "output": 0.3 + "input": 3, + "output": 18, + "reasoning": 18, + "cache_read": 0.3 } }, - "Qwen/Qwen3-VL-235B-A22B-Thinking": { - "id": "Qwen/Qwen3-VL-235B-A22B-Thinking", - "name": "Qwen/Qwen3-VL-235B-A22B-Thinking", - "family": "qwen", - "attachment": true, + "xpersona-frieren-coder": { + "id": "xpersona-frieren-coder", + "name": "Xpersona Frieren 1", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-12-30", + "release_date": "2026-05-01", + "last_updated": "2026-05-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.45, - "output": 3.5 + "input": 1.5, + "output": 6, + "reasoning": 6, + "cache_read": 0.15 } }, - "Qwen/Qwen2.5-7B-Instruct": { - "id": "Qwen/Qwen2.5-7B-Instruct", - "name": "Qwen/Qwen2.5-7B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 3, + "output": 18, + "reasoning": 18, + "cache_read": 0.3 } - }, - "Qwen/Qwen2.5-Coder-32B-Instruct": { - "id": "Qwen/Qwen2.5-Coder-32B-Instruct", - "name": "Qwen/Qwen2.5-Coder-32B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + } + } + }, + "azure-cognitive-services": { + "id": "azure-cognitive-services", + "env": ["AZURE_COGNITIVE_SERVICES_RESOURCE_NAME", "AZURE_COGNITIVE_SERVICES_API_KEY"], + "npm": "@ai-sdk/azure", + "name": "Azure Cognitive Services", + "doc": "https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models", + "models": { + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-11-11", - "last_updated": "2025-11-25", + "knowledge": "2025-03-31", + "release_date": "2025-11-24", + "last_updated": "2025-08-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "Qwen/Qwen3-VL-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-VL-30B-A3B-Instruct", - "name": "Qwen/Qwen3-VL-30B-A3B-Instruct", - "family": "qwen", + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-10-05", - "last_updated": "2025-11-25", + "knowledge": "2025-07-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.29, - "output": 1 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "Qwen/QwQ-32B": { - "id": "Qwen/QwQ-32B", - "name": "Qwen/QwQ-32B", - "family": "qwen", - "attachment": false, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 Nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-03-06", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.58 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "Qwen/Qwen3-235B-A22B": { - "id": "Qwen/Qwen3-235B-A22B", - "name": "Qwen/Qwen3-235B-A22B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "knowledge": "2025-12-31", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.35, - "output": 1.42 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "Qwen/Qwen3-Omni-30B-A3B-Captioner": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Captioner", - "name": "Qwen/Qwen3-Omni-30B-A3B-Captioner", - "family": "qwen", + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "knowledge": "2025-03-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 200000, + "output": 32000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "Qwen/Qwen3-VL-8B-Thinking": { - "id": "Qwen/Qwen3-VL-8B-Thinking", - "name": "Qwen/Qwen3-VL-8B-Thinking", - "family": "qwen", - "attachment": true, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": true, "structured_output": true, "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 0.18, - "output": 2 + "input": 0.6, + "output": 3 } }, - "Qwen/Qwen2.5-VL-7B-Instruct": { - "id": "Qwen/Qwen2.5-VL-7B-Instruct", - "name": "Qwen/Qwen2.5-VL-7B-Instruct", - "family": "qwen", + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2025-01-28", - "last_updated": "2025-11-25", + "knowledge": "2025-02-31", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "Qwen/Qwen3-Next-80B-A3B-Instruct": { - "id": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "name": "Qwen/Qwen3-Next-80B-A3B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 1.4 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "Qwen/Qwen3-8B": { - "id": "Qwen/Qwen3-8B", - "name": "Qwen/Qwen3-8B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 Mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.06, - "output": 0.06 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "Qwen/Qwen3-30B-A3B-Instruct-2507": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "family": "qwen", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": true, "structured_output": true, "temperature": true, - "release_date": "2025-07-30", - "last_updated": "2025-11-25", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/models", + "shape": "completions" }, "cost": { - "input": 0.09, - "output": 0.3 + "input": 0.95, + "output": 4 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-23", - "last_updated": "2025-11-25", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 200000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/anthropic", + "api": "https://${AZURE_COGNITIVE_SERVICES_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" }, "cost": { - "input": 0.09, - "output": 0.6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "Qwen/Qwen3-32B": { - "id": "Qwen/Qwen3-32B", - "name": "Qwen/Qwen3-32B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.4-pro": { + "id": "gpt-5.4-pro", + "name": "GPT-5.4 Pro", + "description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks", + "family": "gpt-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium", "high", "xhigh"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "structured_output": false, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 30, + "output": 180, + "tiers": [ + { + "input": 60, + "output": 270, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 60, + "output": 270 + } } }, - "Qwen/Qwen3-Coder-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", - "name": "Qwen/Qwen3-Coder-30B-A3B-Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-01", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } }, - "Qwen/Qwen3-Omni-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Instruct", - "name": "Qwen/Qwen3-Omni-30B-A3B-Instruct", - "family": "qwen", + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { "input": ["text", "image", "audio"], - "output": ["text"] + "output": ["text", "image", "audio"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "Qwen/Qwen3-14B": { - "id": "Qwen/Qwen3-14B", - "name": "Qwen/Qwen3-14B", - "family": "qwen", + "meta-llama-3-70b-instruct": { + "id": "meta-llama-3-70b-instruct", + "name": "Meta-Llama-3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-04-30", - "last_updated": "2025-11-25", + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 8192, + "output": 2048 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 2.68, + "output": 3.54 } }, - "Qwen/Qwen2.5-72B-Instruct-128K": { - "id": "Qwen/Qwen2.5-72B-Instruct-128K", - "name": "Qwen/Qwen2.5-72B-Instruct-128K", - "family": "qwen", + "phi-3-mini-4k-instruct": { + "id": "phi-3-mini-4k-instruct", + "name": "Phi-3-mini-instruct (4k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 4000 + "context": 4096, + "output": 1024 }, "cost": { - "input": 0.59, - "output": 0.59 + "input": 0.13, + "output": 0.52 } }, - "Qwen/Qwen2.5-32B-Instruct": { - "id": "Qwen/Qwen2.5-32B-Instruct", - "name": "Qwen/Qwen2.5-32B-Instruct", - "family": "qwen", + "phi-3-mini-128k-instruct": { + "id": "phi-3-mini-128k-instruct", + "name": "Phi-3-mini-instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2024-09-19", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 0.13, + "output": 0.52 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "family": "qwen", + "phi-4-mini-reasoning": { + "id": "phi-4-mini-reasoning", + "name": "Phi-4-mini-reasoning", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.13, - "output": 0.6 + "input": 0.075, + "output": 0.3 } }, - "Qwen/Qwen3-Omni-30B-A3B-Thinking": { - "id": "Qwen/Qwen3-Omni-30B-A3B-Thinking", - "name": "Qwen/Qwen3-Omni-30B-A3B-Thinking", - "family": "qwen", - "attachment": true, + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.4 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "Qwen/Qwen2.5-VL-32B-Instruct": { - "id": "Qwen/Qwen2.5-VL-32B-Instruct", - "name": "Qwen/Qwen2.5-VL-32B-Instruct", - "family": "qwen", + "llama-4-maverick-17b-128e-instruct-fp8": { + "id": "llama-4-maverick-17b-128e-instruct-fp8", + "name": "Llama 4 Maverick 17B 128E Instruct FP8", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-03-24", - "last_updated": "2025-11-25", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.27, - "output": 0.27 + "input": 0.25, + "output": 1 } }, - "Qwen/Qwen3-Next-80B-A3B-Thinking": { - "id": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "name": "Qwen/Qwen3-Next-80B-A3B-Thinking", - "family": "qwen", + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek-V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-25", - "last_updated": "2025-11-25", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.58, + "output": 1.68 } }, - "Qwen/Qwen3-VL-235B-A22B-Instruct": { - "id": "Qwen/Qwen3-VL-235B-A22B-Instruct", - "name": "Qwen/Qwen3-VL-235B-A22B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "gpt-5-codex": { + "id": "gpt-5-codex", + "name": "GPT-5-Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "Qwen/Qwen2.5-14B-Instruct": { - "id": "Qwen/Qwen2.5-14B-Instruct", - "name": "Qwen/Qwen2.5-14B-Instruct", - "family": "qwen", - "attachment": false, + "model-router": { + "id": "model-router", + "name": "Model Router", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "family": "model-router", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "release_date": "2025-05-19", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 4000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.14, + "output": 0 } }, - "Qwen/Qwen3-VL-30B-A3B-Thinking": { - "id": "Qwen/Qwen3-VL-30B-A3B-Thinking", - "name": "Qwen/Qwen3-VL-30B-A3B-Thinking", - "family": "qwen", + "gpt-4o-mini": { + "id": "gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Small omni GPT for cheap multimodal assistance and production-scale traffic", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-11", - "last_updated": "2025-11-25", + "knowledge": "2023-09", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.29, - "output": 1 + "input": 0.15, + "output": 0.6, + "cache_read": 0.075 } }, - "Qwen/Qwen3-VL-32B-Instruct": { - "id": "Qwen/Qwen3-VL-32B-Instruct", - "name": "Qwen/Qwen3-VL-32B-Instruct", - "family": "qwen", + "phi-4-multimodal": { + "id": "phi-4-multimodal", + "name": "Phi-4-multimodal", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "phi", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-10-21", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.08, + "output": 0.32, + "input_audio": 4 } }, - "Qwen/Qwen3-VL-8B-Instruct": { - "id": "Qwen/Qwen3-VL-8B-Instruct", - "name": "Qwen/Qwen3-VL-8B-Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "phi-4-reasoning-plus": { + "id": "phi-4-reasoning-plus", + "name": "Phi-4-reasoning-plus", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.18, - "output": 0.68 + "input": 0.125, + "output": 0.5 } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "name": "Qwen/Qwen3-Coder-480B-A35B-Instruct", - "family": "qwen", + "codestral-2501": { + "id": "codestral-2501", + "name": "Codestral 25.01", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "codestral", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-31", - "last_updated": "2025-11-25", + "knowledge": "2024-03", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 256000, + "output": 256000 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.3, + "output": 0.9 } }, - "Qwen/Qwen2.5-72B-Instruct": { - "id": "Qwen/Qwen2.5-72B-Instruct", - "name": "Qwen/Qwen2.5-72B-Instruct", - "family": "qwen", + "cohere-embed-v3-english": { + "id": "cohere-embed-v3-english", + "name": "Embed v3 English", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2025-11-25", + "tool_call": false, + "temperature": false, + "release_date": "2023-11-07", + "last_updated": "2023-11-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 512, + "output": 1024 }, "cost": { - "input": 0.59, - "output": 0.59 + "input": 0.1, + "output": 0 } }, - "stepfun-ai/Step-3.5-Flash": { - "id": "stepfun-ai/Step-3.5-Flash", - "name": "stepfun-ai/Step-3.5-Flash", - "family": "step", + "phi-3-medium-4k-instruct": { + "id": "phi-3-medium-4k-instruct", + "name": "Phi-3-medium-instruct (4k)", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, + "open_weights": true, + "limit": { + "context": 4096, + "output": 1024 + }, + "cost": { + "input": 0.17, + "output": 0.68 + } + }, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.05, + "output": 0.4, + "cache_read": 0.01 } }, - "zai-org/GLM-4.5": { - "id": "zai-org/GLM-4.5", - "name": "zai-org/GLM-4.5", - "family": "glm", - "attachment": false, + "gpt-4-turbo": { + "id": "gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-11-25", + "knowledge": "2023-12", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.4, - "output": 2 + "input": 10, + "output": 30 } }, - "zai-org/GLM-5V-Turbo": { - "id": "zai-org/GLM-5V-Turbo", - "name": "zai-org/GLM-5V-Turbo", - "family": "glm", + "gpt-4.1-mini": { + "id": "gpt-4.1-mini", + "name": "GPT-4.1 mini", + "description": "Affordable GPT-4.1 lane for fast coding help and structured extraction", + "family": "gpt-mini", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 1.2, - "output": 4, - "cache_write": 0 + "input": 0.4, + "output": 1.6, + "cache_read": 0.1 } }, - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", - "name": "zai-org/GLM-4.7", - "family": "glm", - "attachment": false, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] }, - "structured_output": true, + "open_weights": false, + "limit": { + "context": 272000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2, + "cache_read": 0.03 + } + }, + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": true, "temperature": true, - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", - "name": "zai-org/GLM-5.1", - "family": "glm", + "ministral-3b": { + "id": "ministral-3b", + "name": "Ministral 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "knowledge": "2024-03", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 205000, - "output": 205000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_write": 0 + "input": 0.04, + "output": 0.04 } }, - "zai-org/GLM-4.5-Air": { - "id": "zai-org/GLM-4.5-Air", - "name": "zai-org/GLM-4.5-Air", - "family": "glm-air", - "attachment": false, + "llama-3.2-90b-vision-instruct": { + "id": "llama-3.2-90b-vision-instruct", + "name": "Llama-3.2-90B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-11-25", + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.14, - "output": 0.86 + "input": 2.04, + "output": 2.04 } }, - "zai-org/GLM-5": { - "id": "zai-org/GLM-5", - "name": "zai-org/GLM-5", - "family": "glm", + "deepseek-v3.1": { + "id": "deepseek-v3.1", + "name": "DeepSeek-V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2024-07", + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 205000, - "output": 205000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 1, - "output": 3.2 + "input": 0.56, + "output": 1.68 } }, - "zai-org/GLM-4.6V": { - "id": "zai-org/GLM-4.6V", - "name": "zai-org/GLM-4.6V", - "family": "glm", - "attachment": true, + "o1": { + "id": "o1", + "name": "o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2025-12-07", - "last_updated": "2025-12-07", + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-12-05", + "last_updated": "2024-12-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 15, + "output": 60, + "cache_read": 7.5 } }, - "zai-org/GLM-4.6": { - "id": "zai-org/GLM-4.6", - "name": "zai-org/GLM-4.6", - "family": "glm", - "attachment": false, + "gpt-4.1-nano": { + "id": "gpt-4.1-nano", + "name": "GPT-4.1 nano", + "description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks", + "family": "gpt-nano", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-04", - "last_updated": "2025-11-25", + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 205000, - "output": 205000 + "context": 1047576, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 1.9 + "input": 0.1, + "output": 0.4, + "cache_read": 0.025 } }, - "zai-org/GLM-4.5V": { - "id": "zai-org/GLM-4.5V", - "name": "zai-org/GLM-4.5V", - "family": "glm", + "grok-4-fast-reasoning": { + "id": "grok-4-fast-reasoning", + "name": "Grok 4 Fast (Reasoning)", + "description": "Fast Grok model for responsive chat, reasoning, and tool-assisted work", + "family": "grok", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-11-25", + "knowledge": "2025-07", + "release_date": "2025-09-19", + "last_updated": "2025-09-19", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 2000000, + "output": 30000 }, "cost": { - "input": 0.14, - "output": 0.86 + "input": 0.2, + "output": 0.5, + "cache_read": 0.05 } }, - "meta-llama/Meta-Llama-3.1-8B-Instruct": { - "id": "meta-llama/Meta-Llama-3.1-8B-Instruct", - "name": "meta-llama/Meta-Llama-3.1-8B-Instruct", - "family": "llama", + "phi-4": { + "id": "phi-4", + "name": "Phi-4", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-04-23", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 4000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.06, - "output": 0.06 + "input": 0.125, + "output": 0.5 } }, - "inclusionAI/Ring-flash-2.0": { - "id": "inclusionAI/Ring-flash-2.0", - "name": "inclusionAI/Ring-flash-2.0", - "family": "ring", + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "inclusionAI/Ling-mini-2.0": { - "id": "inclusionAI/Ling-mini-2.0", - "name": "inclusionAI/Ling-mini-2.0", - "family": "ling", + "meta-llama-3.1-8b-instruct": { + "id": "meta-llama-3.1-8b-instruct", + "name": "Meta-Llama-3.1-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-10", - "last_updated": "2025-11-25", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.07, - "output": 0.28 + "input": 0.3, + "output": 0.61 } }, - "inclusionAI/Ling-flash-2.0": { - "id": "inclusionAI/Ling-flash-2.0", - "name": "inclusionAI/Ling-flash-2.0", - "family": "ling", + "gpt-3.5-turbo-0301": { + "id": "gpt-3.5-turbo-0301", + "name": "GPT-3.5 Turbo 0301", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "knowledge": "2021-08", + "release_date": "2023-03-01", + "last_updated": "2023-03-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 1.5, + "output": 2 } }, - "tencent/Hunyuan-A13B-Instruct": { - "id": "tencent/Hunyuan-A13B-Instruct", - "name": "tencent/Hunyuan-A13B-Instruct", - "family": "hunyuan", + "deepseek-v3.2-speciale": { + "id": "deepseek-v3.2-speciale", + "name": "DeepSeek-V3.2-Speciale", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-06-30", - "last_updated": "2025-11-25", + "knowledge": "2024-07", + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.58, + "output": 1.68 } }, - "tencent/Hunyuan-MT-7B": { - "id": "tencent/Hunyuan-MT-7B", - "name": "tencent/Hunyuan-MT-7B", - "family": "hunyuan", + "text-embedding-3-small": { + "id": "text-embedding-3-small", + "name": "text-embedding-3-small", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-09-18", - "last_updated": "2025-11-25", + "tool_call": false, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 33000, - "output": 33000 + "context": 8191, + "output": 1536 }, "cost": { - "input": 0, + "input": 0.02, "output": 0 } }, - "deepseek-ai/DeepSeek-V3.1": { - "id": "deepseek-ai/DeepSeek-V3.1", - "name": "deepseek-ai/DeepSeek-V3.1", - "family": "deepseek", + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-25", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 1 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "deepseek-ai/deepseek-vl2": { - "id": "deepseek-ai/deepseek-vl2", - "name": "deepseek-ai/deepseek-vl2", - "family": "deepseek", - "attachment": true, + "mistral-large-2411": { + "id": "mistral-large-2411", + "name": "Mistral Large 24.11", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-13", - "last_updated": "2025-11-25", + "knowledge": "2024-09", + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4000, - "output": 4000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 2, + "output": 6 } }, - "deepseek-ai/DeepSeek-V3": { - "id": "deepseek-ai/DeepSeek-V3", - "name": "deepseek-ai/DeepSeek-V3", - "family": "deepseek", - "attachment": false, + "gpt-4-turbo-vision": { + "id": "gpt-4-turbo-vision", + "name": "GPT-4 Turbo Vision", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2024-12-26", - "last_updated": "2025-11-25", + "knowledge": "2023-11", + "release_date": "2023-11-06", + "last_updated": "2024-04-09", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.25, - "output": 1 + "input": 10, + "output": 30 } }, - "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B": { - "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", - "name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", - "family": "qwen", - "attachment": false, + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.18, - "output": 0.18 + "input": 1.75, + "output": 14, + "cache_read": 0.125 } }, - "deepseek-ai/DeepSeek-R1": { - "id": "deepseek-ai/DeepSeek-R1", - "name": "deepseek-ai/DeepSeek-R1", - "family": "deepseek-thinking", + "o3-mini": { + "id": "o3-mini", + "name": "o3-mini", + "description": "Smaller o-series reasoner for economical coding, math, and planning tasks", + "family": "o-mini", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2024-12-20", + "last_updated": "2025-01-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.5, - "output": 2.18 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B": { - "id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B", - "name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B", - "family": "qwen", + "text-embedding-ada-002": { + "id": "text-embedding-ada-002", + "name": "text-embedding-ada-002", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-01-20", - "last_updated": "2025-11-25", + "reasoning": false, + "tool_call": false, + "release_date": "2022-12-15", + "last_updated": "2022-12-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 8192, + "output": 1536 }, "cost": { "input": 0.1, - "output": 0.1 + "output": 0 } }, - "deepseek-ai/DeepSeek-V3.2-Exp": { - "id": "deepseek-ai/DeepSeek-V3.2-Exp", - "name": "deepseek-ai/DeepSeek-V3.2-Exp", - "family": "deepseek", + "meta-llama-3.1-70b-instruct": { + "id": "meta-llama-3.1-70b-instruct", + "name": "Meta-Llama-3.1-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-10-10", - "last_updated": "2025-11-25", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 164000, - "output": 164000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.27, - "output": 0.41 + "input": 2.68, + "output": 3.54 } }, - "deepseek-ai/DeepSeek-V3.2": { - "id": "deepseek-ai/DeepSeek-V3.2", - "name": "deepseek-ai/DeepSeek-V3.2", - "family": "deepseek", - "attachment": false, - "reasoning": true, + "mistral-medium-2505": { + "id": "mistral-medium-2505", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2025-12-03", + "knowledge": "2025-05", + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 164000, - "output": 164000 + "context": 128000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 0.42 + "input": 0.4, + "output": 2 } }, - "deepseek-ai/DeepSeek-V3.1-Terminus": { - "id": "deepseek-ai/DeepSeek-V3.1-Terminus", - "name": "deepseek-ai/DeepSeek-V3.1-Terminus", - "family": "deepseek", + "phi-4-reasoning": { + "id": "phi-4-reasoning", + "name": "Phi-4-reasoning", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "phi", "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 164000, - "output": 164000 + "context": 32000, + "output": 4096 }, "cost": { - "input": 0.27, - "output": 1 + "input": 0.125, + "output": 0.5 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "openai/gpt-oss-20b", - "family": "gpt-oss", + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt-codex", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "image", "audio"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 8000 + "context": 400000, + "output": 128000 }, "cost": { - "input": 0.04, - "output": 0.18 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "openai/gpt-oss-120b", - "family": "gpt-oss", + "gpt-3.5-turbo-0613": { + "id": "gpt-3.5-turbo-0613", + "name": "GPT-3.5 Turbo 0613", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-11-25", + "knowledge": "2021-08", + "release_date": "2023-06-13", + "last_updated": "2023-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 8000 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.45 + "input": 3, + "output": 4 } }, - "baidu/ERNIE-4.5-300B-A47B": { - "id": "baidu/ERNIE-4.5-300B-A47B", - "name": "baidu/ERNIE-4.5-300B-A47B", - "family": "ernie", + "cohere-embed-v3-multilingual": { + "id": "cohere-embed-v3-multilingual", + "name": "Embed v3 Multilingual", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-02", - "last_updated": "2025-11-25", + "tool_call": false, + "temperature": false, + "release_date": "2023-11-07", + "last_updated": "2023-11-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 512, + "output": 1024 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 0.1, + "output": 0 } }, - "THUDM/GLM-Z1-9B-0414": { - "id": "THUDM/GLM-Z1-9B-0414", - "name": "THUDM/GLM-Z1-9B-0414", - "family": "glm-z", + "gpt-3.5-turbo-0125": { + "id": "gpt-3.5-turbo-0125", + "name": "GPT-3.5 Turbo 0125", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2021-08", + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131000, - "output": 131000 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.086, - "output": 0.086 + "input": 0.5, + "output": 1.5 } }, - "THUDM/GLM-4-9B-0414": { - "id": "THUDM/GLM-4-9B-0414", - "name": "THUDM/GLM-4-9B-0414", - "family": "glm", + "phi-3-small-8k-instruct": { + "id": "phi-3-small-8k-instruct", + "name": "Phi-3-small-instruct (8k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 33000 + "context": 8192, + "output": 2048 }, "cost": { - "input": 0.086, - "output": 0.086 + "input": 0.15, + "output": 0.6 } }, - "THUDM/GLM-4-32B-0414": { - "id": "THUDM/GLM-4-32B-0414", - "name": "THUDM/GLM-4-32B-0414", - "family": "glm", + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": true, - "structured_output": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2024-07", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 33000, - "output": 33000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.27, - "output": 0.27 + "input": 1.35, + "output": 5.4 } }, - "THUDM/GLM-Z1-32B-0414": { - "id": "THUDM/GLM-Z1-32B-0414", - "name": "THUDM/GLM-Z1-32B-0414", - "family": "glm-z", + "kimi-k2-thinking": { + "id": "kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "interleaved": true, "temperature": true, - "release_date": "2025-04-18", - "last_updated": "2025-11-25", + "knowledge": "2024-08", + "release_date": "2025-11-06", + "last_updated": "2025-12-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.14, - "output": 0.57 + "input": 0.6, + "output": 2.5, + "cache_read": 0.15 } }, - "moonshotai/Kimi-K2-Thinking": { - "id": "moonshotai/Kimi-K2-Thinking", - "name": "moonshotai/Kimi-K2-Thinking", - "family": "kimi-thinking", - "attachment": false, + "gpt-5.1-chat": { + "id": "gpt-5.1-chat", + "name": "GPT-5.1 Chat", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-11-07", - "last_updated": "2025-11-25", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-14", + "last_updated": "2025-11-14", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "audio"], + "output": ["text", "image", "audio"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.5 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "moonshotai/Kimi-K2.6", - "family": "kimi", + "llama-3.3-70b-instruct": { + "id": "llama-3.3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.71, + "output": 0.71 } }, - "moonshotai/Kimi-K2-Instruct": { - "id": "moonshotai/Kimi-K2-Instruct", - "name": "moonshotai/Kimi-K2-Instruct", - "family": "kimi", - "attachment": false, + "llama-4-scout-17b-16e-instruct": { + "id": "llama-4-scout-17b-16e-instruct", + "name": "Llama 4 Scout 17B 16E Instruct", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-13", - "last_updated": "2025-11-25", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.58, - "output": 2.29 + "input": 0.2, + "output": 0.78 } }, - "moonshotai/Kimi-K2-Instruct-0905": { - "id": "moonshotai/Kimi-K2-Instruct-0905", - "name": "moonshotai/Kimi-K2-Instruct-0905", - "family": "kimi", + "gpt-3.5-turbo-1106": { + "id": "gpt-3.5-turbo-1106", + "name": "GPT-3.5 Turbo 1106", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, + "tool_call": false, "temperature": true, - "release_date": "2025-09-08", - "last_updated": "2025-11-25", + "knowledge": "2021-08", + "release_date": "2023-11-06", + "last_updated": "2023-11-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 16384, + "output": 16384 }, "cost": { - "input": 0.4, + "input": 1, "output": 2 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "moonshotai/Kimi-K2.5", - "family": "kimi", + "phi-4-mini": { + "id": "phi-4-mini", + "name": "Phi-4-mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2023-10", + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262000, - "output": 262000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.45, - "output": 2.25 + "input": 0.075, + "output": 0.3 } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMaxAI/MiniMax-M2.5", - "family": "minimax", + "cohere-command-r-plus-08-2024": { + "id": "cohere-command-r-plus-08-2024", + "name": "Command R+", + "description": "Cohere's RAG workhorse for long-context enterprise search and tool use", + "family": "command-r", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 197000, - "output": 131000 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 2.5, + "output": 10 } }, - "MiniMaxAI/MiniMax-M2.1": { - "id": "MiniMaxAI/MiniMax-M2.1", - "name": "MiniMaxAI/MiniMax-M2.1", - "family": "minimax", + "meta-llama-3.1-405b-instruct": { + "id": "meta-llama-3.1-405b-instruct", + "name": "Meta-Llama-3.1-405B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "knowledge": "2023-12", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 197000, - "output": 131000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 5.33, + "output": 16 } }, - "ByteDance-Seed/Seed-OSS-36B-Instruct": { - "id": "ByteDance-Seed/Seed-OSS-36B-Instruct", - "name": "ByteDance-Seed/Seed-OSS-36B-Instruct", - "family": "seed", + "gpt-4-32k": { + "id": "gpt-4-32k", + "name": "GPT-4 32K", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-09-04", - "last_updated": "2025-11-25", + "knowledge": "2023-11", + "release_date": "2023-03-14", + "last_updated": "2023-03-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 0.21, - "output": 0.57 + "input": 60, + "output": 120 } - } - } - }, - "vercel": { - "id": "vercel", - "env": ["AI_GATEWAY_API_KEY"], - "npm": "@ai-sdk/gateway", - "name": "Vercel AI Gateway", - "doc": "https://github.com/vercel/ai/tree/5eb85cc45a259553501f535b8ac79a77d0e79223/packages/gateway", - "models": { - "alibaba/qwen3-coder-plus": { - "id": "alibaba/qwen3-coder-plus", - "name": "Qwen3 Coder Plus", - "family": "qwen", + }, + "phi-3-medium-128k-instruct": { + "id": "phi-3-medium-128k-instruct", + "name": "Phi-3-medium-instruct (128k)", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 1000000 + "context": 128000, + "output": 4096 }, "cost": { - "input": 1, - "output": 5 + "input": 0.17, + "output": 0.68 } }, - "alibaba/qwen3.6-27b": { - "id": "alibaba/qwen3.6-27b", - "name": "Qwen 3.6 27B", - "family": "qwen3.6", + "o4-mini": { + "id": "o4-mini", + "name": "o4-mini", + "description": "Fast o-series model for compact reasoning, coding, and tool use", + "family": "o-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-05-01", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.6, - "output": 3.5999999999999996 + "input": 1.1, + "output": 4.4, + "cache_read": 0.275 } }, - "alibaba/qwen3-embedding-8b": { - "id": "alibaba/qwen3-embedding-8b", - "name": "Qwen3 Embedding 8B", - "family": "qwen", + "gpt-4": { + "id": "gpt-4", + "name": "GPT-4", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "tool_call": true, + "temperature": true, + "knowledge": "2023-11", + "release_date": "2023-03-14", + "last_updated": "2023-03-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 8192, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0 + "input": 60, + "output": 120 } }, - "alibaba/qwen-3-30b": { - "id": "alibaba/qwen-3-30b", - "name": "Qwen3-30B-A3B", - "family": "qwen", - "attachment": false, - "reasoning": true, + "gpt-4o": { + "id": "gpt-4o", + "name": "GPT-4o", + "description": "Omni-era GPT for multimodal chat, practical coding, and general assistants", + "family": "gpt", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2023-09", + "release_date": "2024-05-13", + "last_updated": "2024-08-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 40960, + "context": 128000, "output": 16384 }, "cost": { - "input": 0.08, - "output": 0.29 + "input": 2.5, + "output": 10, + "cache_read": 1.25 } }, - "alibaba/qwen-3-235b": { - "id": "alibaba/qwen-3-235b", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", + "cohere-command-r-08-2024": { + "id": "cohere-command-r-08-2024", + "name": "Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2024-06-01", + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 40960, - "output": 16384 + "context": 128000, + "output": 4000 }, "cost": { - "input": 0.13, + "input": 0.15, "output": 0.6 } }, - "alibaba/qwen3.5-flash": { - "id": "alibaba/qwen3.5-flash", - "name": "Qwen 3.5 Flash", - "family": "qwen", + "gpt-5-pro": { + "id": "gpt-5-pro", + "name": "GPT-5 Pro", + "description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-10-06", + "last_updated": "2025-10-06", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 400000, + "output": 272000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.001, - "cache_write": 0.125 + "input": 15, + "output": 120 } }, - "alibaba/qwen3.6-plus": { - "id": "alibaba/qwen3.6-plus", - "name": "Qwen 3.6 Plus", - "family": "qwen", + "llama-3.2-11b-vision-instruct": { + "id": "llama-3.2-11b-vision-instruct", + "name": "Llama-3.2-11B-Vision-Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-03", + "knowledge": "2023-12", + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 128000, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.09999999999999999, - "cache_write": 0.625 + "input": 0.37, + "output": 0.37 } }, - "alibaba/qwen3-max": { - "id": "alibaba/qwen3-max", - "name": "Qwen3 Max", - "family": "qwen", + "cohere-command-a": { + "id": "cohere-command-a", + "name": "Command A", + "description": "Cohere command model for multilingual enterprise agents, tools, and chat", + "family": "command-a", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "knowledge": "2024-06-01", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 256000, + "output": 8000 }, "cost": { - "input": 1.2, - "output": 6 + "input": 2.5, + "output": 10 } }, - "alibaba/qwen3-embedding-0.6b": { - "id": "alibaba/qwen3-embedding-0.6b", - "name": "Qwen3 Embedding 0.6B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5-chat": { + "id": "gpt-5-chat", + "name": "GPT-5 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, "temperature": false, - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "knowledge": "2024-10-24", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 0.01, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "alibaba/qwen-3-32b": { - "id": "alibaba/qwen-3-32b", - "name": "Qwen 3.32B", - "family": "qwen", - "attachment": false, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 40960, - "output": 16384 + "context": 272000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 1.25, + "output": 10, + "cache_read": 0.13 } }, - "alibaba/qwen-3.6-max-preview": { - "id": "alibaba/qwen-3.6-max-preview", - "name": "Qwen 3.6 Max Preview", - "family": "qwen", + "cohere-embed-v-4-0": { + "id": "cohere-embed-v-4-0", + "name": "Embed v4", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "cohere-embed", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-24", + "reasoning": false, + "tool_call": false, + "temperature": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 240000, - "output": 64000 + "context": 128000, + "output": 1536 }, "cost": { - "input": 1.3, - "output": 7.8, - "cache_read": 0.26, - "cache_write": 1.625 + "input": 0.12, + "output": 0 } }, - "alibaba/qwen3-next-80b-a3b-thinking": { - "id": "alibaba/qwen3-next-80b-a3b-thinking", - "name": "Qwen3 Next 80B A3B Thinking", - "family": "qwen", + "mistral-nemo": { + "id": "mistral-nemo", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-12", - "last_updated": "2025-09-12", + "knowledge": "2024-07", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 128000, + "output": 128000 }, "cost": { "input": 0.15, - "output": 1.5 + "output": 0.15 } }, - "alibaba/qwen3-vl-thinking": { - "id": "alibaba/qwen3-vl-thinking", - "name": "Qwen3 VL Thinking", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, + "phi-3.5-mini-instruct": { + "id": "phi-3.5-mini-instruct", + "name": "Phi-3.5-mini-instruct", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 129024 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.7, - "output": 8.4 + "input": 0.13, + "output": 0.52 } }, - "alibaba/qwen3-235b-a22b-thinking": { - "id": "alibaba/qwen3-235b-a22b-thinking", - "name": "Qwen3 235B A22B Thinking 2507", - "family": "qwen", - "attachment": true, + "o1-mini": { + "id": "o1-mini", + "name": "o1-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "temperature": false, + "knowledge": "2023-09", + "release_date": "2024-09-12", + "last_updated": "2024-09-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262114, - "output": 262114 + "context": 128000, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.9 + "input": 1.1, + "output": 4.4, + "cache_read": 0.55 } }, - "alibaba/qwen3-next-80b-a3b-instruct": { - "id": "alibaba/qwen3-next-80b-a3b-instruct", - "name": "Qwen3 Next 80B A3B Instruct", - "family": "qwen", + "text-embedding-3-large": { + "id": "text-embedding-3-large", + "name": "text-embedding-3-large", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "text-embedding", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-12", - "last_updated": "2025-09-12", + "tool_call": false, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 8191, + "output": 3072 }, "cost": { - "input": 0.09, - "output": 1.1 + "input": 0.13, + "output": 0 } }, - "alibaba/qwen3-coder-next": { - "id": "alibaba/qwen3-coder-next", - "name": "Qwen3 Coder Next", - "family": "qwen", - "attachment": false, - "reasoning": true, + "mistral-small-2503": { + "id": "mistral-small-2503", + "name": "Mistral Small 3.1", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "mistral-small", + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-07-22", - "last_updated": "2026-02-19", + "knowledge": "2024-09", + "release_date": "2025-03-01", + "last_updated": "2025-03-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 1.2 + "input": 0.1, + "output": 0.3 } }, - "alibaba/qwen3-embedding-4b": { - "id": "alibaba/qwen3-embedding-4b", - "name": "Qwen3 Embedding 4B", - "family": "qwen", + "meta-llama-3-8b-instruct": { + "id": "meta-llama-3-8b-instruct", + "name": "Meta-Llama-3-8B-Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "temperature": true, + "knowledge": "2023-12", + "release_date": "2024-04-18", + "last_updated": "2024-04-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 8192, + "output": 2048 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.3, + "output": 0.61 } }, - "alibaba/qwen3-max-thinking": { - "id": "alibaba/qwen3-max-thinking", - "name": "Qwen 3 Max Thinking", - "family": "qwen", + "phi-3-small-128k-instruct": { + "id": "phi-3-small-128k-instruct", + "name": "Phi-3-small-instruct (128k)", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "phi", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01", - "last_updated": "2025-01", + "knowledge": "2023-10", + "release_date": "2024-04-23", + "last_updated": "2024-04-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 65536 + "context": 128000, + "output": 4096 }, "cost": { - "input": 1.2, - "output": 6, - "cache_read": 0.24 + "input": 0.15, + "output": 0.6 } }, - "alibaba/qwen3-vl-235b-a22b-instruct": { - "id": "alibaba/qwen3-vl-235b-a22b-instruct", - "name": "Qwen3 VL 235B A22B Instruct", - "family": "qwen", - "attachment": true, + "deepseek-v3-0324": { + "id": "deepseek-v3-0324", + "name": "DeepSeek-V3-0324", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "temperature": true, - "release_date": "2025-09-24", - "last_updated": "2026-05-01", + "knowledge": "2024-07", + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 131072, - "output": 129024 + "output": 131072 }, "cost": { - "input": 0.39999999999999997, - "output": 1.5999999999999999 + "input": 1.14, + "output": 4.56 } }, - "alibaba/qwen3-coder": { - "id": "alibaba/qwen3-coder", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "o3": { + "id": "o3", + "name": "o3", + "description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis", + "family": "o", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "temperature": false, + "knowledge": "2024-05", + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 66536 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.38, - "output": 1.53 + "input": 2, + "output": 8, + "cache_read": 0.5 } }, - "alibaba/qwen3-max-preview": { - "id": "alibaba/qwen3-max-preview", - "name": "Qwen3 Max Preview", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5.2-chat": { + "id": "gpt-5.2-chat", + "name": "GPT-5.2 Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 128000, + "output": 16384 }, "cost": { - "input": 1.2, - "output": 6, - "cache_read": 0.24 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "alibaba/qwen3.5-plus": { - "id": "alibaba/qwen3.5-plus", - "name": "Qwen 3.5 Plus", - "family": "qwen", - "attachment": true, + "deepseek-r1-0528": { + "id": "deepseek-r1-0528", + "name": "DeepSeek-R1-0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-02-16", - "last_updated": "2026-02-19", + "knowledge": "2024-07", + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 163840, + "output": 163840 }, "cost": { - "input": 0.4, - "output": 2.4, - "cache_read": 0.04, - "cache_write": 0.5 + "input": 1.35, + "output": 5.4 } }, - "alibaba/qwen-3-14b": { - "id": "alibaba/qwen-3-14b", - "name": "Qwen3-14B", - "family": "qwen", + "gpt-3.5-turbo-instruct": { + "id": "gpt-3.5-turbo-instruct", + "name": "GPT-3.5 Turbo Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "knowledge": "2021-08", + "release_date": "2023-09-21", + "last_updated": "2023-09-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 40960, - "output": 16384 + "context": 4096, + "output": 4096 }, "cost": { - "input": 0.06, - "output": 0.24 + "input": 1.5, + "output": 2 } }, - "alibaba/qwen3-vl-instruct": { - "id": "alibaba/qwen3-vl-instruct", - "name": "Qwen3 VL Instruct", - "family": "qwen", - "attachment": true, + "phi-3.5-moe-instruct": { + "id": "phi-3.5-moe-instruct", + "name": "Phi-3.5-MoE-instruct", + "description": "Open-weight instruction model for adaptable chat and self-hosted production workloads", + "family": "phi", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-24", - "last_updated": "2025-09-24", + "knowledge": "2023-10", + "release_date": "2024-08-20", + "last_updated": "2024-08-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 129024 + "context": 128000, + "output": 4096 }, "cost": { - "input": 0.7, - "output": 2.8 + "input": 0.16, + "output": 0.64 } }, - "alibaba/qwen3-coder-30b-a3b": { - "id": "alibaba/qwen3-coder-30b-a3b", - "name": "Qwen 3 Coder 30B A3B Instruct", - "family": "qwen", - "attachment": false, + "codex-mini": { + "id": "codex-mini", + "name": "Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "temperature": false, + "knowledge": "2024-04", + "release_date": "2025-05-16", + "last_updated": "2025-05-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 160000, - "output": 32768 + "context": 200000, + "output": 100000 }, "cost": { - "input": 0.07, - "output": 0.27 + "input": 1.5, + "output": 6, + "cache_read": 0.375 } - }, - "perplexity/sonar-pro": { - "id": "perplexity/sonar-pro", - "name": "Sonar Pro", - "family": "sonar-pro", + } + } + }, + "baseten": { + "id": "baseten", + "env": ["BASETEN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://inference.baseten.co/v1", + "name": "Baseten", + "doc": "https://docs.baseten.co/inference/model-apis/overview", + "models": { + "moonshotai/Kimi-K2.6": { + "id": "moonshotai/Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 3, - "output": 15 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "perplexity/sonar": { - "id": "perplexity/sonar", - "name": "Sonar", - "family": "sonar", + "moonshotai/Kimi-K2.5": { + "id": "moonshotai/Kimi-K2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-02", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "knowledge": "2025-12", + "release_date": "2026-01-30", + "last_updated": "2026-02-12", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 127000, - "output": 8000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1, - "output": 1 + "input": 0.6, + "output": 3, + "cache_read": 0.12 } }, - "perplexity/sonar-reasoning": { - "id": "perplexity/sonar-reasoning", - "name": "Sonar Reasoning", - "family": "sonar-reasoning", - "attachment": false, + "moonshotai/Kimi-K2.7-Code": { + "id": "moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 127000, - "output": 8000 + "context": 262000, + "output": 262000 }, "cost": { - "input": 1, - "output": 5 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "perplexity/sonar-reasoning-pro": { - "id": "perplexity/sonar-reasoning-pro", - "name": "Sonar Reasoning Pro", - "family": "sonar-reasoning", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "OpenAI GPT 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "knowledge": "2025-08", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 127000, - "output": 8000 + "context": 128072, + "output": 128072 }, "cost": { - "input": 2, - "output": 8 + "input": 0.1, + "output": 0.5 } }, - "deepseek/deepseek-v3.2-thinking": { - "id": "deepseek/deepseek-v3.2-thinking", - "name": "DeepSeek V3.2 Thinking", - "family": "deepseek-thinking", + "nvidia/Nemotron-120B-A12B": { + "id": "nvidia/Nemotron-120B-A12B", + "name": "Nemotron Super", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2026-02", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 202800, + "output": 202800 }, "cost": { - "input": 0.28, - "output": 0.42, - "cache_read": 0.03 + "input": 0.3, + "output": 0.75, + "cache_read": 0.06 } }, - "deepseek/deepseek-v3.2-exp": { - "id": "deepseek/deepseek-v3.2-exp", - "name": "DeepSeek V3.2 Exp", - "family": "deepseek", + "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B": { + "id": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B", + "name": "Nemotron Ultra", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 163840 + "context": 202800, + "output": 202800 }, "cost": { - "input": 0.27, - "output": 0.4 + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 } }, - "deepseek/deepseek-v3.1": { - "id": "deepseek/deepseek-v3.1", - "name": "DeepSeek-V3.1", - "family": "deepseek", + "zai-org/GLM-5": { + "id": "zai-org/GLM-5", + "name": "GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "knowledge": "2026-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 128000 + "context": 202800, + "output": 202800 }, "cost": { - "input": 0.3, - "output": 1 + "input": 0.95, + "output": 3.15, + "cache_read": 0.2 } }, - "deepseek/deepseek-v4-flash": { - "id": "deepseek/deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek", + "zai-org/GLM-4.7": { + "id": "zai-org/GLM-4.7", + "name": "GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-23", - "last_updated": "2026-04-24", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 200000, + "output": 200000 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.6, + "output": 2.2, + "cache_read": 0.12 } }, - "deepseek/deepseek-v4-pro": { - "id": "deepseek/deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek", + "zai-org/GLM-5.2": { + "id": "zai-org/GLM-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-23", - "last_updated": "2026-04-24", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 1000000, - "output": 384000 + "context": 202720, + "output": 202720 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 1.4, + "output": 4.4, + "cache_read": 0.3 } }, - "deepseek/deepseek-v3.2": { - "id": "deepseek/deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "zai-org/GLM-5.1": { + "id": "zai-org/GLM-5.1", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163842, - "output": 8000 + "context": 202800, + "output": 202800 }, "cost": { - "input": 0.27, - "output": 0.4, - "cache_read": 0.22 + "input": 1.3, + "output": 4.3, + "cache_read": 0.26 } }, - "deepseek/deepseek-v3": { - "id": "deepseek/deepseek-v3", - "name": "DeepSeek V3 0324", + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "DeepSeek V3.1", + "description": "Legacy model retained for compatibility with older integrations", "family": "deepseek", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-12-26", - "last_updated": "2024-12-26", + "release_date": "2025-08-25", + "last_updated": "2025-08-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 163840, - "output": 16384 + "context": 164000, + "output": 131000 }, + "status": "deprecated", "cost": { - "input": 0.77, - "output": 0.77 + "input": 0.5, + "output": 1.5 } }, - "deepseek/deepseek-v3.1-terminus": { - "id": "deepseek/deepseek-v3.1-terminus", - "name": "DeepSeek V3.1 Terminus", - "family": "deepseek", + "deepseek-ai/DeepSeek-V4-Pro": { + "id": "deepseek-ai/DeepSeek-V4-Pro", + "name": "Deepseek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-22", - "last_updated": "2025-09-22", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 65536 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.27, - "output": 1 + "input": 1.74, + "output": 3.48, + "cache_read": 0.145 } }, - "deepseek/deepseek-r1": { - "id": "deepseek/deepseek-r1", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", + "MiniMaxAI/MiniMax-M2.5": { + "id": "MiniMaxAI/MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Legacy model retained for compatibility with older integrations", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-29", + "knowledge": "2026-01", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 204000, + "output": 204000 }, + "status": "deprecated", "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.3, + "output": 1.2 } - }, - "arcee-ai/trinity-mini": { - "id": "arcee-ai/trinity-mini", - "name": "Trinity Mini", - "family": "trinity", + } + } + }, + "atomic-chat": { + "id": "atomic-chat", + "env": ["ATOMIC_CHAT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "http://127.0.0.1:1337/v1", + "name": "Atomic Chat", + "doc": "https://atomic.chat", + "models": { + "gemma-4-E4B-it-IQ4_XS": { + "id": "gemma-4-E4B-it-IQ4_XS", + "name": "Gemma 4 E4B Instruct (IQ4_XS)", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, "reasoning": false, "tool_call": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12", - "last_updated": "2025-12", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131072, - "output": 131072 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.15 + "input": 0, + "output": 0 } }, - "arcee-ai/trinity-large-thinking": { - "id": "arcee-ai/trinity-large-thinking", - "name": "Trinity Large Thinking", - "family": "trinity", + "Meta-Llama-3_1-8B-Instruct-GGUF": { + "id": "Meta-Llama-3_1-8B-Instruct-GGUF", + "name": "Meta Llama 3.1 8B Instruct (GGUF)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-03", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262100, - "output": 80000 + "context": 131072, + "output": 4096 }, "cost": { - "input": 0.25, - "output": 0.8999999999999999 + "input": 0, + "output": 0 } }, - "arcee-ai/trinity-large-preview": { - "id": "arcee-ai/trinity-large-preview", - "name": "Trinity Large Preview", - "family": "trinity", - "attachment": false, + "Qwen3_5-9B-MLX-4bit": { + "id": "Qwen3_5-9B-MLX-4bit", + "name": "Qwen 3.5 9B (MLX 4-bit)", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-01", - "last_updated": "2025-01", + "release_date": "2026-03-05", + "last_updated": "2026-04-04", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 131000, - "output": 131000 + "context": 32768, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0, + "output": 0 } }, - "recraft/recraft-v3": { - "id": "recraft/recraft-v3", - "name": "Recraft V3", - "family": "recraft", + "gemma-4-E4B-it-MLX-4bit": { + "id": "gemma-4-E4B-it-MLX-4bit", + "name": "Gemma 4 E4B Instruct (MLX 4-bit)", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-10", - "last_updated": "2024-10", + "temperature": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 512, + "context": 32768, + "output": 8192 + }, + "cost": { + "input": 0, "output": 0 } }, - "recraft/recraft-v2": { - "id": "recraft/recraft-v2", - "name": "Recraft V2", - "family": "recraft", - "attachment": false, + "Qwen3_5-9B-Q4_K_M": { + "id": "Qwen3_5-9B-Q4_K_M", + "name": "Qwen 3.5 9B (Q4_K_M)", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-03", - "last_updated": "2024-03", + "tool_call": true, + "temperature": true, + "release_date": "2026-03-05", + "last_updated": "2026-04-04", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image"], + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 512, + "context": 32768, + "output": 8192 + }, + "cost": { + "input": 0, "output": 0 } - }, - "voyage/voyage-3-large": { - "id": "voyage/voyage-3-large", - "name": "voyage-3-large", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-09", - "last_updated": "2024-09", + } + } + }, + "meta": { + "id": "meta", + "env": ["META_MODEL_API_KEY"], + "npm": "@ai-sdk/openai", + "api": "https://api.meta.ai/v1", + "name": "Meta", + "doc": "https://dev.meta.ai/docs", + "models": { + "muse-spark-1.1": { + "id": "muse-spark-1.1", + "name": "Muse Spark 1.1", + "description": "Muse Spark is a natively multimodal reasoning model with support for tool-use, visual chain of thought, and multi-agent orchestration.", + "family": "muse", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "minimal", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "release_date": "2026-04-08", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "output": 32000 }, "cost": { - "input": 0.18, - "output": 0 + "input": 1.25, + "output": 4.25, + "cache_read": 0.15 } - }, - "voyage/voyage-4-large": { - "id": "voyage/voyage-4-large", - "name": "voyage-4-large", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, + } + } + }, + "routing-run": { + "id": "routing-run", + "env": ["ROUTING_RUN_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.routing.run/v1", + "name": "routing.run", + "doc": "https://docs.routing.run/api-reference/models", + "models": { + "kimi-k2.6-nitro": { + "id": "kimi-k2.6-nitro", + "name": "Kimi K2.6 Nitro", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-06", - "last_updated": "2026-03-06", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 0 + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 0.275, + "output": 1.1 } }, - "voyage/voyage-3.5-lite": { - "id": "voyage/voyage-3.5-lite", - "name": "voyage-3.5-lite", - "family": "voyage", + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.02, - "output": 0 + "input": 0.112, + "output": 0.224 } }, - "voyage/voyage-code-3": { - "id": "voyage/voyage-code-3", - "name": "voyage-code-3", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "release_date": "2024-09", - "last_updated": "2024-09", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.18, - "output": 0 + "input": 0.275, + "output": 1.1 } }, - "voyage/voyage-finance-2": { - "id": "voyage/voyage-finance-2", - "name": "voyage-finance-2", - "family": "voyage", + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-03", - "last_updated": "2024-03", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.12, - "output": 0 + "input": 0.348, + "output": 0.696 } }, - "voyage/voyage-4-lite": { - "id": "voyage/voyage-4-lite", - "name": "voyage-4-lite", - "family": "voyage", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-06", - "last_updated": "2026-03-06", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 0 + "context": 200000, + "output": 32000 + }, + "cost": { + "input": 0.8, + "output": 2.4 } }, - "voyage/voyage-4": { - "id": "voyage/voyage-4", - "name": "voyage-4", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-06", - "last_updated": "2026-03-06", + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "output": 0 + "context": 1000000, + "output": 32000 + }, + "cost": { + "input": 5, + "output": 25 } }, - "voyage/voyage-code-2": { - "id": "voyage/voyage-code-2", - "name": "voyage-code-2", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "release_date": "2024-01", - "last_updated": "2024-01", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0 + "input": 0.7, + "output": 4.2 } }, - "voyage/voyage-law-2": { - "id": "voyage/voyage-law-2", - "name": "voyage-law-2", - "family": "voyage", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": false, - "release_date": "2024-03", - "last_updated": "2024-03", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.12, - "output": 0 + "input": 1.5, + "output": 9 } }, - "voyage/voyage-3.5": { - "id": "voyage/voyage-3.5", - "name": "voyage-3.5", - "family": "voyage", + "glm-5.2-nitro": { + "id": "glm-5.2-nitro", + "name": "GLM 5.2 Nitro", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.06, - "output": 0 + "input": 0.8, + "output": 2.4 } }, - "morph/morph-v3-large": { - "id": "morph/morph-v3-large", - "name": "Morph v3 Large", - "family": "morph", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, + "context": 200000, "output": 32000 }, "cost": { - "input": 0.9, - "output": 1.9 + "input": 0.275, + "output": 1.1 } }, - "morph/morph-v3-fast": { - "id": "morph/morph-v3-fast", - "name": "Morph v3 Fast", - "family": "morph", + "qwen3.5-9b": { + "id": "qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-08-15", - "last_updated": "2024-08-15", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 16000, - "output": 16000 + "context": 262144, + "output": 32000 }, "cost": { - "input": 0.8, - "output": 1.2 + "input": 0.16, + "output": 0.48 } }, - "zai/glm-5v-turbo": { - "id": "zai/glm-5v-turbo", - "name": "GLM 5V Turbo", - "family": "glm", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-03", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 128000 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24 + "input": 3, + "output": 15 } }, - "zai/glm-4.7": { - "id": "zai/glm-4.7", - "name": "GLM 4.7", - "family": "glm", + "nemotron-3-ultra": { + "id": "nemotron-3-ultra", + "name": "Nemotron 3 Ultra 550B A55B", + "description": "Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "release_date": "2026-06-04", + "last_updated": "2026-06-04", "modalities": { "input": ["text"], "output": ["text"] }, + "open_weights": true, + "limit": { + "context": 131072, + "output": 32000 + }, + "cost": { + "input": 0.1, + "output": 0.1 + } + }, + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, "open_weights": false, "limit": { - "context": 202752, - "output": 120000 + "context": 1000000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.43, - "output": 1.75, - "cache_read": 0.08 + "input": 2.5, + "output": 15 } }, - "zai/glm-5": { - "id": "zai/glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, + "kimi-k2.7-code-nitro": { + "id": "kimi-k2.7-code-nitro", + "name": "Kimi K2.7 Code Nitro", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-19", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202800, - "output": 131072 + "context": 200000, + "output": 32000 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2 + "input": 0.275, + "output": 1.1 } - }, - "zai/glm-4.7-flashx": { - "id": "zai/glm-4.7-flashx", - "name": "GLM 4.7 FlashX", - "family": "glm-flash", + } + } + }, + "aihubmix": { + "id": "aihubmix", + "env": ["AIHUBMIX_API_KEY"], + "npm": "@aihubmix/ai-sdk-provider", + "name": "AIHubMix", + "doc": "https://docs.aihubmix.com", + "models": { + "coding-minimax-m2.7": { + "id": "coding-minimax-m2.7", + "name": "Coding MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-01", - "last_updated": "2025-01", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 128000 + "context": 204800, + "output": 128100 }, "cost": { - "input": 0.06, - "output": 0.4, - "cache_read": 0.01 + "input": 0.2, + "output": 0.2 } }, - "zai/glm-5.1": { - "id": "zai/glm-5.1", - "name": "GLM 5.1", + "alicloud-glm-5.1": { + "id": "alicloud-glm-5.1", + "name": "GLM-5.1 (Alibaba Cloud)", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "family": "glm", - "attachment": true, + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-16", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 202752, - "output": 202752 + "context": 200000, + "output": 128000 }, "cost": { - "input": 1.4, - "output": 4.4, - "cache_read": 0.26 + "input": 0.84, + "output": 3.38, + "cache_read": 0.169, + "cache_write": 1.05625 } }, - "zai/glm-4.6v-flash": { - "id": "zai/glm-4.6v-flash", - "name": "GLM-4.6V-Flash", - "family": "glm", + "claude-sonnet-4-6-think": { + "id": "claude-sonnet-4-6-think", + "name": "Claude Sonnet 4.6 Thinking", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 24000 + "context": 1000000, + "output": 64000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "zai/glm-4.5": { - "id": "zai/glm-4.5", - "name": "GLM 4.5", - "family": "glm", - "attachment": false, + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "cache_write": 1 } }, - "zai/glm-4.5-air": { - "id": "zai/glm-4.5-air", - "name": "GLM 4.5 Air", - "family": "glm-air", + "alicloud-deepseek-v4-flash": { + "id": "alicloud-deepseek-v4-flash", + "name": "DeepSeek V4 Flash (Alibaba Cloud)", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 96000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.2, - "output": 1.1 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "zai/glm-5-turbo": { - "id": "zai/glm-5-turbo", - "name": "GLM 5 Turbo", - "family": "glm", + "xiaomi-mimo-v2.5-pro": { + "id": "xiaomi-mimo-v2.5-pro", + "name": "Xiaomi MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo-v2.5-pro", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-17", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 202800, - "output": 131100 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 1.2, - "output": 4, - "cache_read": 0.24 + "input": 1.1, + "output": 3.3, + "cache_read": 0.22, + "tiers": [ + { + "input": 2.2, + "output": 6.6, + "cache_read": 0.44, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 2.2, + "output": 6.6, + "cache_read": 0.44 + } } }, - "zai/glm-4.5v": { - "id": "zai/glm-4.5v", - "name": "GLM 4.5V", - "family": "glm", + "doubao-seed-2-0-code-preview": { + "id": "doubao-seed-2-0-code-preview", + "name": "Doubao Seed 2.0 Code Preview", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", + "family": "seed", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 66000, - "output": 66000 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 0.48, + "output": 2.41, + "cache_read": 0.09644, + "tiers": [ + { + "input": 0.72, + "output": 3.62, + "cache_read": 0.144656, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 1.45, + "output": 7.23, + "cache_read": 0.28932, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "zai/glm-4.6": { - "id": "zai/glm-4.6", - "name": "GLM 4.6", - "family": "glm", + "coding-xiaomi-mimo-v2.5-pro": { + "id": "coding-xiaomi-mimo-v2.5-pro", + "name": "Coding Xiaomi MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo-v2.5-pro", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 96000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.45, - "output": 1.8 + "input": 0.2, + "output": 0.6, + "cache_read": 0.04, + "tiers": [ + { + "input": 0.4, + "output": 1.2, + "cache_read": 0.08, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.4, + "output": 1.2, + "cache_read": 0.08 + } } }, - "zai/glm-4.6v": { - "id": "zai/glm-4.6v", - "name": "GLM-4.6V", - "family": "glm", + "doubao-seed-2-0-pro": { + "id": "doubao-seed-2-0-pro", + "name": "Doubao Seed 2.0 Pro", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 24000 + "context": 256000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.9, - "cache_read": 0.05 + "input": 0.48, + "output": 2.41, + "cache_read": 0.09644, + "tiers": [ + { + "input": 0.72, + "output": 3.62, + "cache_read": 0.144656, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 1.45, + "output": 7.23, + "cache_read": 0.28932, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "zai/glm-4.7-flash": { - "id": "zai/glm-4.7-flash", - "name": "GLM 4.7 Flash", - "family": "glm", + "deep-deepseek-v4-flash": { + "id": "deep-deepseek-v4-flash", + "name": "DeepSeek V4 Flash (DeepSeek)", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-13", - "last_updated": "2026-03-13", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 131000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.07, - "output": 0.39999999999999997 + "input": 0.154, + "output": 0.308, + "cache_read": 0.0308 } }, - "cohere/command-a": { - "id": "cohere/command-a", - "name": "Command A", - "family": "command", + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Multimodal Qwen workhorse for long-context agents, visual inputs, and coding", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 256000, - "output": 8000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 2.5, - "output": 10 - } - }, - "cohere/embed-v4.0": { - "id": "cohere/embed-v4.0", - "name": "Embed v4.0", - "family": "cohere-embed", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 991000, + "output": 64000 }, "cost": { - "input": 0.12, - "output": 0 + "input": 0.282, + "output": 1.128, + "cache_read": 0.0564, + "cache_write": 0.3525 } }, - "prime-intellect/intellect-3": { - "id": "prime-intellect/intellect-3", - "name": "INTELLECT 3", - "family": "intellect", - "attachment": false, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-11-26", - "last_updated": "2025-11-26", + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 1.1 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "xai/grok-4.3": { - "id": "xai/grok-4.3", + "grok-4.3": { + "id": "grok-4.3", "name": "Grok 4.3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "family": "grok", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-30", + "release_date": "2026-05-01", "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, @@ -94111,882 +146454,1467 @@ "cost": { "input": 1.25, "output": 2.5, - "cache_read": 0.19999999999999998 + "cache_read": 0.2, + "tiers": [ + { + "input": 2.5, + "output": 5, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 5, + "cache_read": 0.4 + } } }, - "xai/grok-4.20-non-reasoning": { - "id": "xai/grok-4.20-non-reasoning", - "name": "Grok 4.20 Non-Reasoning", - "family": "grok", - "attachment": true, - "reasoning": false, + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks", + "family": "qwen", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "max": 262144 + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-23", + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 991000, + "output": 64000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 + "input": 1.69, + "output": 5.07, + "cache_read": 0.169, + "cache_write": 2.1125 } }, - "xai/grok-4.20-non-reasoning-beta": { - "id": "xai/grok-4.20-non-reasoning-beta", - "name": "Grok 4.20 Beta Non-Reasoning", - "family": "grok", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-13", + "knowledge": "2025-01", + "release_date": "2025-03-20", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 } }, - "xai/grok-4.20-reasoning": { - "id": "xai/grok-4.20-reasoning", - "name": "Grok 4.20 Reasoning", - "family": "grok", + "xiaomi-mimo-v2.5-free": { + "id": "xiaomi-mimo-v2.5-free", + "name": "Xiaomi MiMo-V2.5 (free)", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo-v2.5", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-23", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 - } - }, - "xai/grok-imagine-image": { - "id": "xai/grok-imagine-image", - "name": "Grok Imagine Image", - "family": "grok", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-01-28", - "last_updated": "2026-02-19", - "modalities": { - "input": ["text"], - "output": ["text", "image"] - }, - "open_weights": false, - "limit": { - "context": 0, - "output": 0 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "xai/grok-4.20-multi-agent-beta": { - "id": "xai/grok-4.20-multi-agent-beta", - "name": "Grok 4.20 Multi Agent Beta", - "family": "grok", - "attachment": false, + "claude-opus-4-7-think": { + "id": "claude-opus-4-7-think", + "name": "Claude Opus 4.7 Thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-13", + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "xai/grok-imagine-image-pro": { - "id": "xai/grok-imagine-image-pro", - "name": "Grok Imagine Image Pro", - "family": "grok", + "coding-glm-5.1-free": { + "id": "coding-glm-5.1-free", + "name": "Coding GLM 5.1 (free)", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm-free", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-01-28", - "last_updated": "2026-02-19", + "release_date": "2026-04-11", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], - "output": ["text", "image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 0, + "context": 200000, + "output": 128000 + }, + "cost": { + "input": 0, "output": 0 } }, - "xai/grok-4-fast-reasoning": { - "id": "xai/grok-4-fast-reasoning", - "name": "Grok 4 Fast Reasoning", - "family": "grok", + "xiaomi-mimo-v2.5-pro-free": { + "id": "xiaomi-mimo-v2.5-pro-free", + "name": "Xiaomi MiMo-V2.5-Pro (free)", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", + "family": "mimo-v2.5-pro", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 256000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0, + "output": 0, + "cache_read": 0 } }, - "xai/grok-4.1-fast-non-reasoning": { - "id": "xai/grok-4.1-fast-non-reasoning", - "name": "Grok 4.1 Fast Non-Reasoning", - "family": "grok", - "attachment": false, - "reasoning": false, + "xiaomi-mimo-v2.5": { + "id": "xiaomi-mimo-v2.5", + "name": "Xiaomi MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo-v2.5", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.44, + "output": 2.2, + "cache_read": 0.088, + "tiers": [ + { + "input": 0.88, + "output": 4.4, + "cache_read": 0.176, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.88, + "output": 4.4, + "cache_read": 0.176 + } } }, - "xai/grok-4.20-reasoning-beta": { - "id": "xai/grok-4.20-reasoning-beta", - "name": "Grok 4.20 Beta Reasoning", - "family": "grok", + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-13", + "interleaved": true, + "structured_output": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "xai/grok-4.20-multi-agent": { - "id": "xai/grok-4.20-multi-agent", - "name": "Grok 4.20 Multi-Agent", - "family": "grok", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-23", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 2000000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.19999999999999998 + "input": 1.1268, + "output": 3.9438, + "cache_read": 0.2817 } }, - "xai/grok-4.1-fast-reasoning": { - "id": "xai/grok-4.1-fast-reasoning", - "name": "Grok 4.1 Fast Reasoning", - "family": "grok", - "attachment": false, + "qwen3.6-flash": { + "id": "qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.6", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "knowledge": "2025-04", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 991000, + "output": 64000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.17, + "output": 1.01, + "cache_read": 0.0169, + "cache_write": 0.21125, + "tiers": [ + { + "input": 0.68, + "output": 4.06, + "cache_read": 0.0676, + "cache_write": 0.845, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.68, + "output": 4.06, + "cache_read": 0.0676, + "cache_write": 0.845 + } } }, - "xai/grok-4-fast-non-reasoning": { - "id": "xai/grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", + "coding-xiaomi-mimo-v2.5": { + "id": "coding-xiaomi-mimo-v2.5", + "name": "Coding Xiaomi MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", + "family": "mimo-v2.5", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-05-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.08, + "output": 0.4, + "cache_read": 0.016, + "tiers": [ + { + "input": 0.16, + "output": 0.8, + "cache_read": 0.032, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 0.16, + "output": 0.8, + "cache_read": 0.032 + } } }, - "xai/grok-3": { - "id": "xai/grok-3", - "name": "Grok 3", - "family": "grok", - "attachment": false, - "reasoning": false, + "gpt-5.1-codex": { + "id": "gpt-5.1-codex", + "name": "GPT-5.1 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "xai/grok-3-mini": { - "id": "xai/grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", - "attachment": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "interleaved": true, + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.3, - "output": 0.5, - "reasoning": 0.5, - "cache_read": 0.075 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "xai/grok-2-vision": { - "id": "xai/grok-2-vision", - "name": "Grok 2 Vision", - "family": "grok", + "gemini-3.1-pro-preview-customtools": { + "id": "gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-08", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { "input": 2, - "output": 10, - "cache_read": 2 + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "xai/grok-4": { - "id": "xai/grok-4", - "name": "Grok 4", - "family": "grok", - "attachment": false, + "doubao-seed-2-0-mini-260428": { + "id": "doubao-seed-2-0-mini-260428", + "name": "Doubao Seed 2.0 Mini 260428", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "output": 64000 + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "reasoning": 15, - "cache_read": 0.75 + "input": 0.03, + "output": 0.28, + "cache_read": 0.00564, + "input_audio": 0.423, + "tiers": [ + { + "input": 0.06, + "output": 0.56, + "cache_read": 0.01128, + "input_audio": 0.846, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 0.11, + "output": 1.13, + "cache_read": 0.02256, + "input_audio": 1.692, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "xai/grok-code-fast-1": { - "id": "xai/grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", - "attachment": false, + "doubao-seed-2-0-lite-260428": { + "id": "doubao-seed-2-0-lite-260428", + "name": "Doubao Seed 2.0 Lite 260428", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "seed", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "output": 10000 + "output": 128000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.08, + "output": 0.51, + "cache_read": 0.01692, + "input_audio": 1.269, + "tiers": [ + { + "input": 0.13, + "output": 0.76, + "cache_read": 0.02536, + "input_audio": 1.902, + "tier": { + "type": "context", + "size": 32000 + } + }, + { + "input": 0.25, + "output": 1.52, + "cache_read": 0.05072, + "input_audio": 3.804, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "xai/grok-3-fast": { - "id": "xai/grok-3-fast", - "name": "Grok 3 Fast", - "family": "grok", - "attachment": false, - "reasoning": false, + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 1.25 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "xai/grok-3-mini-fast": { - "id": "xai/grok-3-mini-fast", - "name": "Grok 3 Mini Fast", - "family": "grok", - "attachment": false, + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 4, - "reasoning": 4, - "cache_read": 0.15 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "nvidia/nemotron-3-super-120b-a12b": { - "id": "nvidia/nemotron-3-super-120b-a12b", - "name": "NVIDIA Nemotron 3 Super 120B A12B", - "family": "nemotron", + "minimax-m2.7": { + "id": "minimax-m2.7", + "name": "MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, "release_date": "2026-03-18", - "last_updated": "2026-03-30", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 32000 + "context": 204800, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.65 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "nvidia/nemotron-3-nano-30b-a3b": { - "id": "nvidia/nemotron-3-nano-30b-a3b", - "name": "Nemotron 3 Nano 30B A3B", - "family": "nemotron", - "attachment": false, + "claude-opus-4-6-think": { + "id": "claude-opus-4-6-think", + "name": "Claude Opus 4.6 Thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12", - "last_updated": "2024-12", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.06, - "output": 0.24 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "nvidia/nemotron-nano-12b-v2-vl": { - "id": "nvidia/nemotron-nano-12b-v2-vl", - "name": "Nvidia Nemotron Nano 12B V2 VL", - "family": "nemotron", + "gpt-5.1-codex-mini": { + "id": "gpt-5.1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12", - "last_updated": "2024-12", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "nvidia/nemotron-nano-9b-v2": { - "id": "nvidia/nemotron-nano-9b-v2", - "name": "Nvidia Nemotron Nano 9B V2", - "family": "nemotron", - "attachment": false, + "kimi-k2.5": { + "id": "kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-18", - "last_updated": "2025-08-18", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 131072, - "output": 131072 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.04, - "output": 0.16 - } - }, - "inception/mercury-edit-2": { - "id": "inception/mercury-edit-2", - "name": "Mercury Edit 2", - "attachment": false, - "reasoning": true, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-30", - "last_updated": "2026-03-30", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.025 + "input": 0.6, + "output": 3, + "cache_read": 0.1 } }, - "inception/mercury-2": { - "id": "inception/mercury-2", - "name": "Mercury 2", - "family": "mercury", + "coding-glm-5.1": { + "id": "coding-glm-5.1", + "name": "Coding GLM 5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-03-06", + "release_date": "2026-04-11", + "last_updated": "2026-04-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, + "context": 200000, "output": 128000 }, "cost": { - "input": 0.25, - "output": 0.75, - "cache_read": 0.024999999999999998 + "input": 0.06, + "output": 0.22, + "cache_read": 0.013 } }, - "inception/mercury-coder-small": { - "id": "inception/mercury-coder-small", - "name": "Mercury Coder Small Beta", - "family": "mercury", + "coding-minimax-m2.7-free": { + "id": "coding-minimax-m2.7-free", + "name": "Coding MiniMax M2.7 (Free)", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax-free", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-02-26", - "last_updated": "2026-05-01", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 32000, - "output": 16384 + "context": 204800, + "output": 128100 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0, + "output": 0 } }, - "openai/gpt-5.1-codex-max": { - "id": "openai/gpt-5.1-codex-max", - "name": "GPT 5.1 Codex Max", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1050000, + "input": 922000, "output": 128000 }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 5, + "output": 30, + "cache_read": 0.5 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } + }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "openai/gpt-5.2-chat": { - "id": "openai/gpt-5.2-chat", - "name": "GPT-5.2 Chat", - "family": "gpt", + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.18 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "openai/gpt-4o-mini-search-preview": { - "id": "openai/gpt-4o-mini-search-preview", - "name": "GPT 4o Mini Search Preview", - "family": "gpt-mini", - "attachment": false, - "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2025-01", - "last_updated": "2025-01", + "kimi-k2.6": { + "id": "kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.95, + "output": 4, + "cache_read": 0.16 } }, - "openai/codex-mini": { - "id": "openai/codex-mini", - "name": "Codex Mini", - "family": "gpt-codex-mini", + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-16", - "last_updated": "2025-05-16", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "input": 100000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.5, - "output": 6, - "cache_read": 0.38 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "openai/gpt-5-chat": { - "id": "openai/gpt-5-chat", - "name": "GPT-5 Chat", - "family": "gpt", + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "interleaved": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "openai/gpt-5.3-chat": { - "id": "openai/gpt-5.3-chat", - "name": "GPT-5.3 Chat", - "family": "gpt", - "attachment": true, + "deep-deepseek-v4-pro": { + "id": "deep-deepseek-v4-pro", + "name": "DeepSeek V4 Pro (DeepSeek)", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-03", - "last_updated": "2026-03-06", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.478, + "output": 0.956, + "cache_read": 0.004302 } }, - "openai/gpt-5.2-pro": { - "id": "openai/gpt-5.2-pro", - "name": "GPT 5.2 ", - "family": "gpt", + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM 5 Vision Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2026-05-09", + "last_updated": "2026-05-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 200000, "output": 128000 }, "cost": { - "input": 21, - "output": 168 + "input": 0.7042, + "output": 3.09848, + "cache_read": 0.169008 } }, - "openai/text-embedding-3-large": { - "id": "openai/text-embedding-3-large", - "name": "text-embedding-3-large", - "family": "text-embedding", + "zai-glm-5.1": { + "id": "zai-glm-5.1", + "name": "GLM-5.1 (Z.ai)", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "input": 6656, - "output": 1536 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.13, - "output": 0 + "input": 0.845, + "output": 3.38, + "cache_read": 0.183112 } }, - "openai/gpt-5.5": { - "id": "openai/gpt-5.5", - "name": "GPT 5.5", - "family": "gpt", + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, + "interleaved": true, + "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -94994,155 +147922,221 @@ "open_weights": false, "limit": { "context": 1000000, - "input": 872000, - "output": 128000 + "output": 64000 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "openai/gpt-5.3-codex": { - "id": "openai/gpt-5.3-codex", - "name": "GPT 5.3 Codex", - "family": "gpt", + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "tiers": [ + { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 0.5, + "output": 3, + "cache_read": 0.05 + } } }, - "openai/text-embedding-ada-002": { - "id": "openai/text-embedding-ada-002", - "name": "text-embedding-ada-002", - "family": "text-embedding", + "coding-minimax-m2.7-highspeed": { + "id": "coding-minimax-m2.7-highspeed", + "name": "Coding MiniMax M2.7 Highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2022-12-15", - "last_updated": "2022-12-15", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 8192, - "input": 6656, - "output": 1536 - }, - "cost": { - "input": 0.1, - "output": 0 - } - }, - "openai/gpt-5.2": { - "id": "openai/gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 204800, + "output": 128100 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.18 + "input": 0.2, + "output": 0.2 } }, - "openai/o3-pro": { - "id": "openai/o3-pro", - "name": "o3 Pro", - "family": "o-pro", + "claude-opus-4-8-think": { + "id": "claude-opus-4-8-think", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": false, - "knowledge": "2024-10", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "input": 100000, - "output": 100000 + "output": 32000 }, "cost": { - "input": 20, - "output": 80 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "openai/gpt-5.4-mini": { - "id": "openai/gpt-5.4-mini", - "name": "GPT 5.4 Mini", - "family": "gpt", - "attachment": true, + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "qwen3.6", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "knowledge": "2025-04", + "release_date": "2026-05-09", + "last_updated": "2026-05-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 240000, + "output": 64000 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 1.27, + "output": 7.61, + "cache_read": 0.1268, + "cache_write": 1.585, + "tiers": [ + { + "input": 2.11, + "output": 12.67, + "cache_read": 0.2112, + "cache_write": 2.64, + "tier": { + "type": "context", + "size": 128000 + } + } + ] } }, - "openai/gpt-5.4-nano": { - "id": "openai/gpt-5.4-nano", - "name": "GPT 5.4 Nano", - "family": "gpt", + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -95154,79 +148148,124 @@ "output": 128000 }, "cost": { - "input": 0.19999999999999998, - "output": 1.25, - "cache_read": 0.02 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "openai/gpt-5.2-codex": { - "id": "openai/gpt-5.2-codex", - "name": "GPT-5.2-Codex", - "family": "gpt-codex", + "qwen3.6-plus": { + "id": "qwen3.6-plus", + "name": "Qwen3.6 Plus", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "family": "qwen3.6", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12", - "last_updated": "2025-12", + "knowledge": "2025-04", + "release_date": "2026-05-09", + "last_updated": "2026-05-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 991000, + "output": 64000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.28, + "output": 1.69, + "cache_read": 0.0282, + "cache_write": 0.3525, + "tiers": [ + { + "input": 1.13, + "output": 6.77, + "cache_read": 0.1128, + "cache_write": 1.41, + "tier": { + "type": "context", + "size": 256000 + } + } + ], + "context_over_200k": { + "input": 1.13, + "output": 6.77, + "cache_read": 0.1128, + "cache_write": 1.41 + } } }, - "openai/gpt-5.1-codex-mini": { - "id": "openai/gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex mini", - "family": "gpt", - "attachment": true, + "alicloud-deepseek-v4-pro": { + "id": "alicloud-deepseek-v4-pro", + "name": "DeepSeek V4 Pro (Alibaba Cloud)", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-16", - "last_updated": "2025-05-16", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.03 + "input": 1.69, + "output": 3.38, + "cache_read": 0.13 } }, - "openai/gpt-5.1-thinking": { - "id": "openai/gpt-5.1-thinking", - "name": "GPT 5.1 Thinking", + "gpt-5.1": { + "id": "gpt-5.1", + "name": "GPT-5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": false, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { @@ -95240,16 +148279,25 @@ "cache_read": 0.13 } }, - "openai/gpt-5.4-pro": { - "id": "openai/gpt-5.4-pro", - "name": "GPT 5.4 Pro", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-03-05", - "last_updated": "2026-03-06", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -95260,1058 +148308,1460 @@ "input": 922000, "output": 128000 }, - "cost": { - "input": 30, - "output": 180 - } - }, - "openai/gpt-3.5-turbo": { - "id": "openai/gpt-3.5-turbo", - "name": "GPT-3.5 Turbo", - "family": "gpt", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2021-09", - "release_date": "2023-03-01", - "last_updated": "2023-03-01", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 16385, - "input": 12289, - "output": 4096 - }, - "cost": { - "input": 0.5, - "output": 1.5 - } - }, - "openai/o3-deep-research": { - "id": "openai/o3-deep-research", - "name": "o3-deep-research", - "family": "o", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-10", - "release_date": "2024-06-26", - "last_updated": "2024-06-26", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "input": 100000, - "output": 100000 + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 12.5, + "output": 75, + "cache_read": 1.25 + }, + "provider": { + "body": { + "service_tier": "priority" + } + } + } + } }, "cost": { - "input": 10, - "output": 40, - "cache_read": 2.5 + "input": 5, + "output": 30, + "cache_read": 0.5, + "tiers": [ + { + "input": 10, + "output": 45, + "cache_read": 1, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 45, + "cache_read": 1 + } } - }, - "openai/text-embedding-3-small": { - "id": "openai/text-embedding-3-small", - "name": "text-embedding-3-small", - "family": "text-embedding", + } + } + }, + "google-vertex": { + "id": "google-vertex", + "env": ["GOOGLE_VERTEX_PROJECT", "GOOGLE_VERTEX_LOCATION", "GOOGLE_APPLICATION_CREDENTIALS"], + "npm": "@ai-sdk/google-vertex", + "name": "Vertex", + "doc": "https://cloud.google.com/vertex-ai/generative-ai/docs/models", + "models": { + "gemini-2.5-pro-tts": { + "id": "gemini-2.5-pro-tts", + "name": "Gemini 2.5 Pro TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gemini-pro", "attachment": false, "reasoning": false, "tool_call": false, "temperature": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "knowledge": "2025-01", + "release_date": "2025-09-30", + "last_updated": "2025-12-10", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 6656, - "output": 1536 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.02, - "output": 0 + "input": 1, + "output": 20 } }, - "openai/gpt-5.4": { - "id": "openai/gpt-5.4", - "name": "GPT 5.4", - "family": "gpt", + "claude-haiku-4-5@20251001": { + "id": "claude-haiku-4-5@20251001", + "name": "Claude Haiku 4.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-03-05", - "last_updated": "2026-03-06", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, + "gemini-3.1-flash-lite": { + "id": "gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-01", + "release_date": "2026-05-07", + "last_updated": "2026-05-07", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "input": 98304, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.3 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 } }, - "openai/gpt-5-pro": { - "id": "openai/gpt-5-pro", - "name": "GPT-5 pro", - "family": "gpt", + "gemini-3.1-flash-image": { + "id": "gemini-3.1-flash-image", + "name": "Nano Banana 2", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", + "family": "gemini-flash", "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], + "tool_call": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "pdf"], "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 128000, - "output": 272000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 15, - "output": 120 + "input": 0.5, + "output": 60 } }, - "openai/gpt-oss-safeguard-20b": { - "id": "openai/gpt-oss-safeguard-20b", - "name": "gpt-oss-safeguard-20b", - "family": "gpt-oss", - "attachment": false, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "input": 65536, + "context": 1048576, "output": 65536 }, "cost": { - "input": 0.08, - "output": 0.3, - "cache_read": 0.04 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "gemini-2.5-flash-tts": { + "id": "gemini-2.5-flash-tts", + "name": "Gemini 2.5 Flash TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "gemini-flash", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning": false, + "tool_call": false, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2025-09-30", + "last_updated": "2025-12-10", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 32768, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.5, + "output": 10 } }, - "openai/gpt-5.5-pro": { - "id": "openai/gpt-5.5-pro", - "name": "GPT 5.5 Pro", - "family": "gpt", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "input": 872000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 30, - "output": 180 + "input": 0.3, + "output": 2.5, + "cache_read": 0.075, + "cache_write": 0.383 } }, - "openai/gpt-3.5-turbo-instruct": { - "id": "openai/gpt-3.5-turbo-instruct", - "name": "GPT-3.5 Turbo Instruct", - "family": "gpt", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2021-09", - "release_date": "2023-03-01", - "last_updated": "2023-03-01", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "input": 4096, - "output": 4096 + "context": 1048576, + "output": 65536 }, "cost": { "input": 1.5, - "output": 2 + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 } }, - "openai/gpt-5.1-instant": { - "id": "openai/gpt-5.1-instant", - "name": "GPT-5.1 Instant", - "family": "gpt", + "claude-opus-4@20250514": { + "id": "claude-opus-4@20250514", + "name": "Claude Opus 4", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 111616, - "output": 16384 + "context": 200000, + "output": 32000 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "openai/gpt-5.1-codex": { - "id": "openai/gpt-5.1-codex", - "name": "GPT-5.1-Codex", - "family": "gpt", + "claude-opus-4-1@20250805": { + "id": "claude-opus-4-1@20250805", + "name": "Claude Opus 4.1", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "output": 32000 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", - "attachment": true, + "gemini-embedding-001": { + "id": "gemini-embedding-001", + "name": "Gemini Embedding 001", + "description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines", + "family": "gemini", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "temperature": false, + "knowledge": "2025-05", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 2048, + "output": 1 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.15, + "output": 0 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "claude-opus-4-5@20251101": { + "id": "claude-opus-4-5@20251101", + "name": "Claude Opus 4.5", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-05", + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 - }, - "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 - } - }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", - "modalities": { - "input": ["text", "image"], - "output": ["text"] + "context": 200000, + "output": 64000 }, - "open_weights": false, - "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "GPT-4o", - "family": "gpt", + "claude-3-5-haiku@20241022": { + "id": "claude-3-5-haiku@20241022", + "name": "Claude Haiku 3.5", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "family": "claude-haiku", "attachment": true, "reasoning": false, "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "knowledge": "2024-07-31", + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "output": 8192 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.8, + "output": 4, + "cache_read": 0.08, + "cache_write": 1 } }, - "openai/o3": { - "id": "openai/o3", - "name": "o3", - "family": "o", + "gemini-3.1-pro-preview-customtools": { + "id": "gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { "input": 2, - "output": 8, - "cache_read": 0.5 + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "openai/gpt-4.1-nano": { - "id": "openai/gpt-4.1-nano", - "name": "GPT-4.1 nano", - "family": "gpt-nano", + "gemini-flash-lite-latest": { + "id": "gemini-flash-lite-latest", + "name": "Gemini Flash-Lite Latest", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { "input": 0.1, "output": 0.4, - "cache_read": 0.03 + "cache_read": 0.025 } }, - "openai/gpt-5-codex": { - "id": "openai/gpt-5-codex", - "name": "GPT-5-Codex", - "family": "gpt-codex", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "gemini-2.5-flash-image": { + "id": "gemini-2.5-flash-image", + "name": "Nano Banana", + "description": "Nano Banana image model for fast generation, edits, and character-consistent assets", + "family": "gemini-flash", + "attachment": true, + "reasoning": false, + "tool_call": false, + "temperature": true, + "knowledge": "2024-06", + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 32768, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 30 } }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "o3-mini", - "family": "o-mini", - "attachment": false, + "claude-sonnet-4@20250514": { + "id": "claude-sonnet-4@20250514", + "name": "Claude Sonnet 4", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 100000 + "output": 64000 + }, + "status": "deprecated", + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openai/o1": { - "id": "openai/o1", - "name": "o1", - "family": "o", + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash-Lite", + "description": "Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents", + "family": "gemini-flash-lite", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 512, + "max": 24576 + } + ], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.1, + "output": 0.4, + "cache_read": 0.01, + "input_audio": 0.3 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "o4-mini", - "family": "o-mini", + "claude-opus-4-7@default": { + "id": "claude-opus-4-7@default", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "openai/gpt-4-turbo": { - "id": "openai/gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", + "claude-sonnet-4-5@20250929": { + "id": "claude-sonnet-4-5@20250929", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 200000, + "output": 64000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 10, - "output": 30 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", + "claude-sonnet-5@default": { + "id": "claude-sonnet-5@default", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1000000, "output": 128000 }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" + }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 } }, - "amazon/titan-embed-text-v2": { - "id": "amazon/titan-embed-text-v2", - "name": "Titan Text Embeddings V2", - "family": "titan-embed", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-04", - "last_updated": "2024-04", + "claude-opus-4-6@default": { + "id": "claude-opus-4-6@default", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], + "tool_call": true, + "temperature": true, + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 0.02, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "amazon/nova-2-lite": { - "id": "amazon/nova-2-lite", - "name": "Nova 2 Lite", - "family": "nova", + "gemini-flash-latest": { + "id": "gemini-flash-latest", + "name": "Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, - "tool_call": false, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-01", - "last_updated": "2024-12-01", + "knowledge": "2025-01", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 1000000 + "context": 1048576, + "output": 65536 }, "cost": { "input": 0.3, - "output": 2.5 + "output": 2.5, + "cache_read": 0.075, + "cache_write": 0.383 } }, - "amazon/nova-pro": { - "id": "amazon/nova-pro", - "name": "Nova Pro", - "family": "nova-pro", + "claude-opus-4-8@default": { + "id": "claude-opus-4-8@default", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 8192 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 0.8, - "output": 3.2, - "cache_read": 0.2 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25, + "tiers": [ + { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 10, + "output": 37.5, + "cache_read": 1, + "cache_write": 12.5 + } } }, - "amazon/nova-lite": { - "id": "amazon/nova-lite", - "name": "Nova Lite", - "family": "nova-lite", + "gemini-3-pro-image": { + "id": "gemini-3-pro-image", + "name": "Nano Banana Pro", + "description": "Nano Banana Pro for higher-fidelity image generation and design-heavy edits", + "family": "gemini-pro", "attachment": true, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "knowledge": "2025-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 300000, - "output": 8192 + "context": 65536, + "output": 32768 }, "cost": { - "input": 0.06, - "output": 0.24, - "cache_read": 0.015 + "input": 2, + "output": 120 } }, - "amazon/nova-micro": { - "id": "amazon/nova-micro", - "name": "Nova Micro", - "family": "nova-micro", - "attachment": false, - "reasoning": false, + "claude-sonnet-4-6@default": { + "id": "claude-sonnet-4-6@default", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-03", - "last_updated": "2024-12-03", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1000000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/google-vertex/anthropic" }, "cost": { - "input": 0.035, - "output": 0.14, - "cache_read": 0.00875 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75, + "tiers": [ + { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 6, + "output": 22.5, + "cache_read": 0.6, + "cache_write": 7.5 + } } }, - "mistral/mistral-nemo": { - "id": "mistral/mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, + "gemini-3.1-flash-lite-preview": { + "id": "gemini-3.1-flash-lite-preview", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 60288, - "output": 16000 + "context": 1048576, + "output": 65536 }, + "status": "deprecated", "cost": { - "input": 0.04, - "output": 0.17 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 } }, - "mistral/ministral-14b": { - "id": "mistral/ministral-14b", - "name": "Ministral 14B", - "family": "ministral", - "attachment": true, - "reasoning": false, - "tool_call": false, + "moonshotai/kimi-k2-thinking-maas": { + "id": "moonshotai/kimi-k2-thinking-maas", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2024-08", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 262144, + "output": 262144 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.6, + "output": 2.5 } }, - "mistral/codestral-embed": { - "id": "mistral/codestral-embed", - "name": "Codestral Embed", - "family": "codestral-embed", + "openai/gpt-oss-120b-maas": { + "id": "openai/gpt-oss-120b-maas", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "temperature": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0 + "input": 0.09, + "output": 0.36 } }, - "mistral/mistral-medium": { - "id": "mistral/mistral-medium", - "name": "Mistral Medium 3.1", - "family": "mistral-medium", - "attachment": true, - "reasoning": false, + "openai/gpt-oss-20b-maas": { + "id": "openai/gpt-oss-20b-maas", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.07, + "output": 0.25 } }, - "mistral/mistral-embed": { - "id": "mistral/mistral-embed", - "name": "Mistral Embed", - "family": "mistral-embed", + "zai-org/glm-4.7-maas": { + "id": "zai-org/glm-4.7-maas", + "name": "GLM-4.7", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-12-11", - "last_updated": "2023-12-11", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2026-01-06", + "last_updated": "2026-01-06", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 1536 + "context": 200000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.6, + "output": 2.2 } }, - "mistral/devstral-2": { - "id": "mistral/devstral-2", - "name": "Devstral 2", - "family": "devstral", + "zai-org/glm-5-maas": { + "id": "zai-org/glm-5-maas", + "name": "GLM-5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 202752, + "output": 131072 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + }, + "cost": { + "input": 1, + "output": 3.2, + "cache_read": 0.1 } }, - "mistral/mistral-large-3": { - "id": "mistral/mistral-large-3", - "name": "Mistral Large 3", - "family": "mistral-large", - "attachment": true, - "reasoning": false, - "tool_call": false, + "deepseek-ai/deepseek-v3.1-maas": { + "id": "deepseek-ai/deepseek-v3.1-maas", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2025-08-28", + "last_updated": "2025-08-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 163840, + "output": 32768 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.6, + "output": 1.7 } }, - "mistral/devstral-small-2": { - "id": "mistral/devstral-small-2", - "name": "Devstral Small 2", - "family": "devstral", + "deepseek-ai/deepseek-v3.2-maas": { + "id": "deepseek-ai/deepseek-v3.2-maas", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "release_date": "2025-12-17", + "last_updated": "2026-04-04", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 256000, - "output": 256000 + "context": 163840, + "output": 65536 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + }, + "cost": { + "input": 0.56, + "output": 1.68, + "cache_read": 0.056 } }, - "mistral/devstral-small": { - "id": "mistral/devstral-small", - "name": "Devstral Small 1.1", - "family": "devstral", + "qwen/qwen3-235b-a22b-instruct-2507-maas": { + "id": "qwen/qwen3-235b-a22b-instruct-2507-maas", + "name": "Qwen3 235B A22B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "release_date": "2025-08-13", + "last_updated": "2025-08-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 64000 + "context": 262144, + "output": 16384 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.22, + "output": 0.88 } }, - "mistral/ministral-8b": { - "id": "mistral/ministral-8b", - "name": "Ministral 8B (latest)", - "family": "ministral", + "meta/llama-3.3-70b-instruct-maas": { + "id": "meta/llama-3.3-70b-instruct-maas", + "name": "Llama 3.3 70B Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-04", + "knowledge": "2023-12", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { "input": ["text"], "output": ["text"] @@ -96319,299 +149769,343 @@ "open_weights": true, "limit": { "context": 128000, - "output": 128000 + "output": 8192 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.72, + "output": 0.72 } }, - "mistral/magistral-medium": { - "id": "mistral/magistral-medium", - "name": "Magistral Medium (latest)", - "family": "magistral-medium", - "attachment": false, - "reasoning": true, + "meta/llama-4-maverick-17b-128e-instruct-maas": { + "id": "meta/llama-4-maverick-17b-128e-instruct-maas", + "name": "Llama 4 Maverick 17B 128E Instruct", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-03-17", - "last_updated": "2025-03-20", + "knowledge": "2024-08", + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 524288, + "output": 8192 + }, + "provider": { + "npm": "@ai-sdk/openai-compatible", + "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" }, "cost": { - "input": 2, - "output": 5 + "input": 0.35, + "output": 1.15 } - }, - "mistral/mistral-small": { - "id": "mistral/mistral-small", - "name": "Mistral Small (latest)", - "family": "mistral-small", + } + } + }, + "nano-gpt": { + "id": "nano-gpt", + "env": ["NANO_GPT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://nano-gpt.com/api/v1", + "name": "NanoGPT", + "doc": "https://docs.nano-gpt.com", + "models": { + "step-3": { + "id": "step-3", + "name": "Step-3", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2026-03-16", - "last_updated": "2026-03-16", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-31", + "last_updated": "2025-07-31", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 65536, + "input": 65536, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.2499, + "output": 0.6494 } }, - "mistral/magistral-small": { - "id": "mistral/magistral-small", - "name": "Magistral Small", - "family": "magistral-small", - "attachment": false, + "qwen3.5-35b-a3b:thinking": { + "id": "qwen3.5-35b-a3b:thinking", + "name": "Qwen3.5 35B A3B Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-03-17", - "last_updated": "2025-03-17", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 260096, + "input": 260096, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.225, + "output": 1.8 } }, - "mistral/pixtral-12b": { - "id": "mistral/pixtral-12b", - "name": "Pixtral 12B", - "family": "pixtral", + "glm-4.1v-thinking-flashx": { + "id": "glm-4.1v-thinking-flashx", + "name": "GLM 4.1V Thinking FlashX", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-09-01", - "last_updated": "2024-09-01", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 64000, + "input": 64000, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.3, + "output": 0.3 } }, - "mistral/mixtral-8x22b-instruct": { - "id": "mistral/mixtral-8x22b-instruct", - "name": "Mixtral 8x22B", - "family": "mixtral", - "attachment": false, + "ernie-x1.1-preview": { + "id": "ernie-x1.1-preview", + "name": "ERNIE X1.1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-17", - "last_updated": "2024-04-17", + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-10", + "last_updated": "2025-09-10", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 64000, - "output": 64000 + "input": 64000, + "output": 8192 }, "cost": { - "input": 2, - "output": 6 + "input": 0.15, + "output": 0.6 } }, - "mistral/pixtral-large": { - "id": "mistral/pixtral-large", - "name": "Pixtral Large (latest)", - "family": "pixtral", + "qwen25-vl-72b-instruct": { + "id": "qwen25-vl-72b-instruct", + "name": "Qwen25 VL 72b", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2024-11-04", + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-10", + "last_updated": "2025-05-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 32000, + "input": 32000, + "output": 32768 }, "cost": { - "input": 2, - "output": 6 + "input": 0.69989, + "output": 0.69989 } }, - "mistral/ministral-3b": { - "id": "mistral/ministral-3b", - "name": "Ministral 3B (latest)", - "family": "ministral", - "attachment": false, + "gemini-2.0-pro-exp-02-05": { + "id": "gemini-2.0-pro-exp-02-05", + "name": "Gemini 2.0 Pro 0205", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-04", + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-05", + "last_updated": "2025-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 2097152, + "input": 2097152, + "output": 8192 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 1.989, + "output": 7.956 } }, - "mistral/codestral": { - "id": "mistral/codestral", - "name": "Codestral (latest)", - "family": "codestral", + "doubao-seed-2-0-lite-260215": { + "id": "doubao-seed-2-0-lite-260215", + "name": "Doubao Seed 2.0 Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-05-29", - "last_updated": "2025-01-04", + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 256000, - "output": 4096 + "input": 256000, + "output": 32000 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.1462, + "output": 0.8738 } }, - "meta/llama-3.2-1b": { - "id": "meta/llama-3.2-1b", - "name": "Llama 3.2 1B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "Qwen3.5-27B-Writer-V2-Derestricted": { + "id": "Qwen3.5-27B-Writer-V2-Derestricted", + "name": "Qwen3.5 27B Writer V2 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.306, + "output": 0.306 } }, - "meta/llama-3.1-8b": { - "id": "meta/llama-3.1-8b", - "name": "Llama 3.1 8B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "claude-opus-4-thinking:8192": { + "id": "claude-opus-4-thinking:8192", + "name": "Claude 4 Opus Thinking (8K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.03, - "output": 0.05 + "input": 14.994, + "output": 75.004 } }, - "meta/llama-3.2-90b": { - "id": "meta/llama-3.2-90b", - "name": "Llama 3.2 90B Vision Instruct", - "family": "llama", + "qwen3-vl-235b-a22b-thinking": { + "id": "qwen3-vl-235b-a22b-thinking", + "name": "Qwen3 VL 235B A22B Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.72, - "output": 0.72 + "input": 0.5, + "output": 6 } }, - "meta/llama-3.2-3b": { - "id": "meta/llama-3.2-3b", - "name": "Llama 3.2 3B Instruct", - "family": "llama", + "glm-4-air-0111": { + "id": "glm-4-air-0111", + "name": "GLM 4 Air 0111", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "structured_output": false, + "release_date": "2025-01-11", + "last_updated": "2025-01-11", "modalities": { "input": ["text"], "output": ["text"] @@ -96619,404 +150113,524 @@ "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 0.1394, + "output": 0.1394 } }, - "meta/llama-3.2-11b": { - "id": "meta/llama-3.2-11b", - "name": "Llama 3.2 11B Vision Instruct", - "family": "llama", + "Qwen3.5-27B-Queen-Derestricted": { + "id": "Qwen3.5-27B-Queen-Derestricted", + "name": "Qwen3.5 27B Queen Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 0.306, + "output": 0.306 + } + }, + "gemini-2.5-pro-preview-03-25": { + "id": "gemini-2.5-pro-preview-03-25", + "name": "Gemini 2.5 Pro Preview 0325", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.16, - "output": 0.16 + "input": 2.5, + "output": 10 } }, - "meta/llama-3.1-70b": { - "id": "meta/llama-3.1-70b", - "name": "Llama 3.1 70B Instruct", - "family": "llama", + "brave-research": { + "id": "brave-research", + "name": "Brave (Research)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": false, + "structured_output": false, + "release_date": "2023-03-02", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, + "context": 16384, + "input": 16384, "output": 16384 }, "cost": { - "input": 0.4, - "output": 0.4 + "input": 5, + "output": 5 } }, - "meta/llama-3.3-70b": { - "id": "meta/llama-3.3-70b", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "qwen-plus": { + "id": "qwen-plus", + "name": "Qwen Plus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-25", + "last_updated": "2024-01-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 995904, + "input": 995904, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3995, + "output": 1.2002 } }, - "meta/llama-4-maverick": { - "id": "meta/llama-4-maverick", - "name": "Llama-4-Maverick-17B-128E-Instruct-FP8", - "family": "llama", - "attachment": true, + "doubao-seed-2-0-mini-260215": { + "id": "doubao-seed-2-0-mini-260215", + "name": "Doubao Seed 2.0 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 256000, + "input": 256000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.0493, + "output": 0.4845 } }, - "meta/llama-4-scout": { - "id": "meta/llama-4-scout", - "name": "Llama-4-Scout-17B-16E-Instruct-FP8", - "family": "llama", + "Qwen3.5-27B-BlueStar-v3-Derestricted-Lite": { + "id": "Qwen3.5-27B-BlueStar-v3-Derestricted-Lite", + "name": "Qwen3.5 27B BlueStar v3 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 0.306, + "output": 0.306 + } + }, + "deepseek-v3-0324": { + "id": "deepseek-v3-0324", + "name": "DeepSeek Chat 0324", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "structured_output": true, + "release_date": "2025-03-24", + "last_updated": "2025-03-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "input": 128000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 0.7 } }, - "vercel/v0-1.5-md": { - "id": "vercel/v0-1.5-md", - "name": "v0-1.5-md", - "family": "v0", + "Gemma-4-31B-Musica-v1": { + "id": "Gemma-4-31B-Musica-v1", + "name": "Gemma 4 31B Musica v1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-06-09", - "last_updated": "2025-06-09", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 3, - "output": 15 + "input": 0.306, + "output": 0.306 } }, - "vercel/v0-1.0-md": { - "id": "vercel/v0-1.0-md", - "name": "v0-1.0-md", - "family": "v0", - "attachment": true, + "mirothinker-1-7-deepresearch-mini": { + "id": "mirothinker-1-7-deepresearch-mini", + "name": "MiroThinker 1.7 Deep Research Mini", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 1.25, + "output": 10 + } + }, + "Baichuan4-Turbo": { + "id": "Baichuan4-Turbo", + "name": "Baichuan 4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", + "modalities": { + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 32000 + "input": 128000, + "output": 32768 }, "cost": { - "input": 3, - "output": 15 + "input": 2.42, + "output": 2.42 } }, - "minimax/minimax-m2.7": { - "id": "minimax/minimax-m2.7", - "name": "Minimax M2.7", - "family": "minimax", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "doubao-seed-1-6-flash-250615": { + "id": "doubao-seed-1-6-flash-250615", + "name": "Doubao Seed 1.6 Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-15", + "last_updated": "2025-06-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131000 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.0374, + "output": 0.374 } }, - "minimax/minimax-m2.7-highspeed": { - "id": "minimax/minimax-m2.7-highspeed", - "name": "MiniMax M2.7 High Speed", - "family": "minimax", + "kimi-k2-instruct-fast": { + "id": "kimi-k2-instruct-fast", + "name": "Kimi K2 0711 Fast", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-15", + "last_updated": "2025-07-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131100 + "context": 131072, + "input": 131072, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.1, + "output": 2 } }, - "minimax/minimax-m2": { - "id": "minimax/minimax-m2", - "name": "MiniMax M2", - "family": "minimax", + "glm-4-plus": { + "id": "glm-4-plus", + "name": "GLM-4 Plus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-08-01", + "last_updated": "2024-08-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262114, - "output": 262114 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.27, - "output": 1.15, - "cache_read": 0.03, - "cache_write": 0.38 + "input": 7.497, + "output": 7.497 } }, - "minimax/minimax-m2.1": { - "id": "minimax/minimax-m2.1", - "name": "MiniMax M2.1", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "v0-1.0-md": { + "id": "v0-1.0-md", + "name": "v0 1.0 MD", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-04", + "last_updated": "2025-07-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "input": 200000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.38 + "input": 3, + "output": 15 } }, - "minimax/minimax-m2.1-lightning": { - "id": "minimax/minimax-m2.1-lightning", - "name": "MiniMax M2.1 Lightning", - "family": "minimax", + "qwen3-coder-30b-a3b-instruct": { + "id": "qwen3-coder-30b-a3b-instruct", + "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "structured_output": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.4, - "cache_read": 0.03, - "cache_write": 0.38 + "input": 0.1, + "output": 0.4 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "gemini-2.5-flash-preview-09-2025-thinking": { + "id": "gemini-2.5-flash-preview-09-2025-thinking", + "name": "Gemini 2.5 Flash Preview (09/2025) – Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-19", + "structured_output": true, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, - "output": 131000 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "output": 2.5 } }, - "minimax/minimax-m2.5-highspeed": { - "id": "minimax/minimax-m2.5-highspeed", - "name": "MiniMax M2.5 High Speed", - "family": "minimax", + "v0-1.5-lg": { + "id": "v0-1.5-lg", + "name": "v0 1.5 LG", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-03-13", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-04", + "last_updated": "2025-07-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 15, + "output": 75 } }, - "kwaipilot/kat-coder-pro-v1": { - "id": "kwaipilot/kat-coder-pro-v1", - "name": "KAT-Coder-Pro V1", - "family": "kat-coder", - "attachment": false, + "qwen3.5-flash:thinking": { + "id": "qwen3.5-flash:thinking", + "name": "Qwen3.5 Flash Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-10-24", - "last_updated": "2025-10-24", + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 32000 + "context": 991808, + "input": 991808, + "output": 65536 + }, + "cost": { + "input": 0.09, + "output": 0.36 } }, - "kwaipilot/kat-coder-pro-v2": { - "id": "kwaipilot/kat-coder-pro-v2", - "name": "Kat Coder Pro V2", - "family": "kat-coder", + "jamba-large": { + "id": "jamba-large", + "name": "Jamba Large", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-30", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text"], "output": ["text"] @@ -97024,2880 +150638,3116 @@ "open_weights": false, "limit": { "context": 256000, - "output": 256000 + "input": 256000, + "output": 4096 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 1.989, + "output": 7.99 } }, - "google/gemini-2.5-flash-lite-preview-09-2025": { - "id": "google/gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview 09-25", - "family": "gemini-flash-lite", + "kimi-thinking-preview": { + "id": "kimi-thinking-preview", + "name": "Kimi Thinking Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-07", + "last_updated": "2025-05-07", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.01 + "input": 31.46, + "output": 31.46 } }, - "google/gemini-3.1-flash-lite-preview": { - "id": "google/gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini", + "gemini-2.0-flash-thinking-exp-01-21": { + "id": "gemini-2.0-flash-thinking-exp-01-21", + "name": "Gemini 2.0 Flash Thinking 0121", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-03", - "last_updated": "2026-03-06", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-21", + "last_updated": "2025-01-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 65000 + "input": 1000000, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "cache_write": 1 + "input": 0.306, + "output": 1.003 } }, - "google/gemini-3-pro-image": { - "id": "google/gemini-3-pro-image", - "name": "Nano Banana Pro (Gemini 3 Pro Image)", - "family": "gemini-pro", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-09", - "last_updated": "2025-09", + "claude-opus-4-1-thinking:1024": { + "id": "claude-opus-4-1-thinking:1024", + "name": "Claude 4.1 Opus Thinking (1K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], - "output": ["text", "image"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 32768 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 2, - "output": 120 + "input": 14.994, + "output": 75.004 } }, - "google/gemini-3.1-pro-preview": { - "id": "google/gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini", + "Qwen3.5-27B-NaNovel-Derestricted-Lite": { + "id": "Qwen3.5-27B-NaNovel-Derestricted-Lite", + "name": "Qwen3.5 27B NaNovel Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-02-19", - "last_updated": "2026-02-24", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2 + "input": 0.306, + "output": 0.306 } }, - "google/gemini-3-pro-preview": { - "id": "google/gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "doubao-seed-1-6-250615": { + "id": "doubao-seed-1-6-250615", + "name": "Doubao Seed 1.6", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-15", + "last_updated": "2025-06-15", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.204, + "output": 0.51 } }, - "google/imagen-4.0-ultra-generate-001": { - "id": "google/imagen-4.0-ultra-generate-001", - "name": "Imagen 4 Ultra", - "family": "imagen", - "attachment": false, - "reasoning": false, + "ernie-5.0-thinking-preview": { + "id": "ernie-5.0-thinking-preview", + "name": "Ernie 5.0 Thinking Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": false, - "release_date": "2025-05-24", - "last_updated": "2025-05-24", + "structured_output": false, + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 128000, + "input": 128000, + "output": 16384 + }, + "cost": { + "input": 1.1, + "output": 2 } }, - "google/gemini-embedding-001": { - "id": "google/gemini-embedding-001", - "name": "Gemini Embedding 001", - "family": "gemini-embedding", - "attachment": false, - "reasoning": false, + "gemini-2.5-flash-preview-05-20:thinking": { + "id": "gemini-2.5-flash-preview-05-20:thinking", + "name": "Gemini 2.5 Flash 0520 Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": false, + "structured_output": false, "release_date": "2025-05-20", "last_updated": "2025-05-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 1048000, + "input": 1048000, + "output": 65536 }, "cost": { "input": 0.15, - "output": 0 + "output": 3.5 } }, - "google/gemma-4-31b-it": { - "id": "google/gemma-4-31b-it", - "name": "Gemma 4 31B IT", - "family": "gemma", + "Qwen3.5-27B-Omega-Evolution-v2.2-Derestricted": { + "id": "Qwen3.5-27B-Omega-Evolution-v2.2-Derestricted", + "name": "Qwen3.5 27B Omega Evolution v2.2 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-03", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 262144, - "output": 131072 + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.39999999999999997 + "input": 0.306, + "output": 0.306 } }, - "google/gemini-2.5-flash-image": { - "id": "google/gemini-2.5-flash-image", - "name": "Nano Banana (Gemini 2.5 Flash Image)", - "family": "gemini-flash", + "azure-o1": { + "id": "azure-o1", + "name": "Azure o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-03-20", + "structured_output": false, + "release_date": "2024-12-17", + "last_updated": "2024-12-17", "modalities": { "input": ["text"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 14.994, + "output": 59.993 } }, - "google/text-embedding-005": { - "id": "google/text-embedding-005", - "name": "Text Embedding 005", - "family": "text-embedding", - "attachment": false, + "qwen3.7-plus": { + "id": "qwen3.7-plus", + "name": "Qwen3.7 Plus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-08", - "last_updated": "2024-08", + "structured_output": false, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 991808, + "input": 991808, + "output": 65536 }, "cost": { - "input": 0.03, - "output": 0 + "input": 0.4, + "output": 1.6, + "cache_read": 0.04 } }, - "google/text-multilingual-embedding-002": { - "id": "google/text-multilingual-embedding-002", - "name": "Text Multilingual Embedding 002", - "family": "text-embedding", + "glm-4-airx": { + "id": "glm-4-airx", + "name": "GLM-4 AirX", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-03", - "last_updated": "2024-03", + "structured_output": false, + "release_date": "2024-06-05", + "last_updated": "2024-06-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 8000, + "input": 8000, + "output": 4096 }, "cost": { - "input": 0.03, - "output": 0 + "input": 2.006, + "output": 2.006 } }, - "google/gemini-3.1-flash-image-preview": { - "id": "google/gemini-3.1-flash-image-preview", - "name": "Gemini 3.1 Flash Image Preview (Nano Banana 2)", - "family": "gemini", - "attachment": true, - "reasoning": true, + "step-r1-v-mini": { + "id": "step-r1-v-mini", + "name": "Step R1 V Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2026-02-26", - "last_updated": "2026-03-06", + "structured_output": false, + "release_date": "2025-04-08", + "last_updated": "2025-04-08", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 3 + "input": 2.5, + "output": 11 } }, - "google/gemini-3.1-flash-lite": { - "id": "google/gemini-3.1-flash-lite", - "name": "Gemini 3.1 Flash Lite", - "family": "gemini", + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-05-07", - "last_updated": "2026-05-08", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65000 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.03 + "input": 2.5, + "output": 10 } }, - "google/gemini-3-flash": { - "id": "google/gemini-3-flash", - "name": "Gemini 3 Flash", - "family": "gemini-flash", + "claude-haiku-4-5-20251001-thinking": { + "id": "claude-haiku-4-5-20251001-thinking", + "name": "Claude Haiku 4.5 Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "structured_output": true, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 200000, + "input": 200000, "output": 64000 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05 + "input": 1, + "output": 5, + "cache_read": 0.1 } }, - "google/imagen-4.0-generate-001": { - "id": "google/imagen-4.0-generate-001", - "name": "Imagen 4", - "family": "imagen", + "jamba-mini-1.6": { + "id": "jamba-mini-1.6", + "name": "Jamba Mini 1.6", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": false, + "release_date": "2025-03-01", + "last_updated": "2025-03-01", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 256000, + "input": 256000, + "output": 4096 + }, + "cost": { + "input": 0.1989, + "output": 0.408 } }, - "google/gemini-2.5-flash-preview-09-2025": { - "id": "google/gemini-2.5-flash-preview-09-2025", - "name": "Gemini 2.5 Flash Preview 09-25", - "family": "gemini-flash", + "claude-haiku-4-5-20251001": { + "id": "claude-haiku-4-5-20251001", + "name": "Claude Haiku 4.5", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "structured_output": true, + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 64000 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "cache_write": 0.383 + "input": 1, + "output": 5 } }, - "google/gemini-embedding-2": { - "id": "google/gemini-embedding-2", - "name": "Gemini Embedding 2", - "family": "gemini-embedding", - "attachment": false, + "claude-3-5-haiku-20241022": { + "id": "claude-3-5-haiku-20241022", + "name": "Claude 3.5 Haiku", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2026-03-10", - "last_updated": "2026-03-23", + "tool_call": true, + "structured_output": true, + "release_date": "2024-10-22", + "last_updated": "2024-10-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 0, - "output": 0 + "context": 200000, + "input": 200000, + "output": 8192 + }, + "cost": { + "input": 0.8, + "output": 4 } }, - "google/gemma-4-26b-a4b-it": { - "id": "google/gemma-4-26b-a4b-it", - "name": "Gemma 4 26B A4B IT", - "family": "gemma", + "claude-sonnet-4-thinking:32768": { + "id": "claude-sonnet-4-thinking:32768", + "name": "Claude 4 Sonnet Thinking (32K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-03", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 0.13, - "output": 0.39999999999999997 + "input": 2.992, + "output": 14.994 } }, - "google/imagen-4.0-fast-generate-001": { - "id": "google/imagen-4.0-fast-generate-001", - "name": "Imagen 4 Fast", - "family": "imagen", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2025-06", - "last_updated": "2025-06", + "hermes-low": { + "id": "hermes-low", + "name": "Hermes Low", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 480, - "output": 0 + "context": 1048576, + "input": 1048576, + "output": 65536 + }, + "cost": { + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 } }, - "google/gemini-2.5-flash-lite": { - "id": "google/gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", + "claude-sonnet-4-thinking": { + "id": "claude-sonnet-4-thinking", + "name": "Claude 4 Sonnet Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "structured_output": true, + "release_date": "2025-02-24", + "last_updated": "2025-02-24", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.01 + "input": 2.992, + "output": 14.994 } }, - "google/gemini-2.5-flash-image-preview": { - "id": "google/gemini-2.5-flash-image-preview", - "name": "Nano Banana Preview (Gemini 2.5 Flash Image Preview)", - "family": "gemini-flash", + "qwen3.7-max": { + "id": "qwen3.7-max", + "name": "Qwen3.7 Max", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-03-20", + "structured_output": false, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { "input": ["text"], - "output": ["text", "image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1000000, + "input": 1000000, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5 + "input": 2.5, + "output": 7.5, + "cache_read": 0.25 } }, - "google/gemini-2.0-flash-lite": { - "id": "google/gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, + "yi-medium-200k": { + "id": "yi-medium-200k", + "name": "Yi Medium 200k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "tool_call": false, + "structured_output": false, + "release_date": "2024-03-01", + "last_updated": "2024-03-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 4096 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 2.499, + "output": 2.499 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", + "Qwen3.5-27B-BlueStar-v2-Derestricted-Lite": { + "id": "Qwen3.5-27B-BlueStar-v2-Derestricted-Lite", + "name": "Qwen3.5 27B BlueStar v2 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "input_audio": 1 + "input": 0.306, + "output": 0.306 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + "Gemma-4-31B-Claude-4.6-Opus-Reasoning-Distilled": { + "id": "Gemma-4-31B-Claude-4.6-Opus-Reasoning-Distilled", + "name": "Gemma 4 31B Claude 4.6 Opus Reasoning Distilled", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - } + "input": 0.306, + "output": 0.306, + "cache_read": 0.0306 } }, - "google/gemini-2.0-flash": { - "id": "google/gemini-2.0-flash", - "name": "Gemini 2.0 Flash", - "family": "gemini-flash", + "doubao-1-5-thinking-vision-pro-250428": { + "id": "doubao-1-5-thinking-vision-pro-250428", + "name": "Doubao 1.5 Thinking Vision Pro", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-15", + "last_updated": "2025-05-15", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.55, + "output": 1.43 } }, - "moonshotai/kimi-k2-turbo": { - "id": "moonshotai/kimi-k2-turbo", - "name": "Kimi K2 Turbo", - "family": "kimi", + "Qwen3.5-27B-BlueStar-Derestricted-Lite": { + "id": "Qwen3.5-27B-BlueStar-Derestricted-Lite", + "name": "Qwen3.5 27B BlueStar Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 0.306, + "output": 0.306 + } + }, + "Qwen2.5-32B-EVA-v0.2": { + "id": "Qwen2.5-32B-EVA-v0.2", + "name": "Qwen 2.5 32b EVA", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "tool_call": false, + "structured_output": false, + "release_date": "2024-09-01", + "last_updated": "2024-09-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 24576, + "input": 24576, + "output": 8192 }, "cost": { - "input": 2.4, - "output": 10 + "input": 0.493, + "output": 0.493 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "gemini-2.5-flash": { + "id": "gemini-2.5-flash", + "name": "Gemini 2.5 Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-26", - "last_updated": "2026-01-26", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 1.2 + "input": 0.3, + "output": 2.5 } }, - "moonshotai/kimi-k2-thinking-turbo": { - "id": "moonshotai/kimi-k2-thinking-turbo", - "name": "Kimi K2 Thinking Turbo", - "family": "kimi-thinking", - "attachment": false, + "qwen-long": { + "id": "qwen-long", + "name": "Qwen Long 10M", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-25", + "last_updated": "2025-01-25", + "modalities": { + "input": ["text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 10000000, + "input": 10000000, + "output": 8192 + }, + "cost": { + "input": 0.1003, + "output": 0.408 + } + }, + "Gemma-4-31B-DarkIdol": { + "id": "Gemma-4-31B-DarkIdol", + "name": "Gemma 4 31B DarkIdol", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262114, - "output": 262114 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1.15, - "output": 8, - "cache_read": 0.15 + "input": 0.306, + "output": 0.306 } }, - "moonshotai/kimi-k2-0905": { - "id": "moonshotai/kimi-k2-0905", - "name": "Kimi K2 0905", - "family": "kimi", + "step-2-16k-exp": { + "id": "step-2-16k-exp", + "name": "Step-2 16k Exp", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "structured_output": false, + "release_date": "2024-07-05", + "last_updated": "2024-07-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 16000, + "input": 16000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 7.004, + "output": 19.992 } }, - "moonshotai/kimi-k2.6": { - "id": "moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "Qwen3.5-27B-Marvin-V2-Derestricted-Lite": { + "id": "Qwen3.5-27B-Marvin-V2-Derestricted-Lite", + "name": "Qwen3.5 27B Marvin V2 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-24", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262000, - "output": 262000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.306, + "output": 0.306 } }, - "moonshotai/kimi-k2-thinking": { - "id": "moonshotai/kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "Qwen3.5-27B-NaNovel-Derestricted": { + "id": "Qwen3.5-27B-NaNovel-Derestricted", + "name": "Qwen3.5 27B NaNovel Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, - "limit": { - "context": 216144, - "output": 216144 + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.47, - "output": 2, - "cache_read": 0.14 + "input": 0.306, + "output": 0.306 } }, - "moonshotai/kimi-k2": { - "id": "moonshotai/kimi-k2", - "name": "Kimi K2 Instruct", - "family": "kimi", + "ernie-4.5-8k-preview": { + "id": "ernie-4.5-8k-preview", + "name": "Ernie 4.5 8k Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-14", - "last_updated": "2025-07-14", + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 8000, + "input": 8000, "output": 16384 }, - "status": "deprecated", "cost": { - "input": 1, - "output": 3 + "input": 0.66, + "output": 2.6 } }, - "interfaze/interfaze-beta": { - "id": "interfaze/interfaze-beta", - "name": "Interfaze Beta", + "gemini-2.0-flash-exp-image-generation": { + "id": "gemini-2.0-flash-exp-image-generation", + "name": "Gemini Text + Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2025-10-07", - "last_updated": "2026-04-29", + "structured_output": false, + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32000 + "context": 32767, + "input": 32767, + "output": 8192 }, "cost": { - "input": 1.5, - "output": 3.5 + "input": 0.2, + "output": 0.8 } }, - "anthropic/claude-3.5-sonnet-20240620": { - "id": "anthropic/claude-3.5-sonnet-20240620", - "name": "Claude 3.5 Sonnet (2024-06-20)", - "family": "claude-sonnet", + "gemini-2.5-flash-lite-preview-09-2025": { + "id": "gemini-2.5-flash-lite-preview-09-2025", + "name": "Gemini 2.5 Flash Lite Preview (09/2025)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-06-20", - "last_updated": "2024-06-20", + "structured_output": true, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 3, - "output": 15 + "input": 0.1, + "output": 0.4 } }, - "anthropic/claude-opus-4.6": { - "id": "anthropic/claude-opus-4.6", - "name": "Claude Opus 4.6", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02", - "last_updated": "2026-02", + "deepseek-reasoner-cheaper": { + "id": "deepseek-reasoner-cheaper", + "name": "Deepseek R1 Cheaper", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.4, + "output": 1.7 } }, - "anthropic/claude-opus-4.7": { - "id": "anthropic/claude-opus-4.7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "venice-uncensored": { + "id": "venice-uncensored", + "name": "Venice Uncensored", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-24", + "last_updated": "2025-02-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.4, + "output": 0.4 } }, - "anthropic/claude-opus-4.5": { - "id": "anthropic/claude-opus-4.5", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "gemini-2.0-flash-001": { + "id": "gemini-2.0-flash-001", + "name": "Gemini 2.0 Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "structured_output": true, + "release_date": "2024-12-11", + "last_updated": "2024-12-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "input": 1000000, + "output": 8192 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 18.75 + "input": 0.1003, + "output": 0.408 } }, - "anthropic/claude-haiku-4.5": { - "id": "anthropic/claude-haiku-4.5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "gemma-4-31B-Larkspur-v0.5": { + "id": "gemma-4-31B-Larkspur-v0.5", + "name": "Gemma 4 31B Larkspur v0.5", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.306, + "output": 0.306 } }, - "anthropic/claude-sonnet-4.6": { - "id": "anthropic/claude-sonnet-4.6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "claude-opus-4-1-20250805": { + "id": "claude-opus-4-1-20250805", + "name": "Claude 4.1 Opus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "structured_output": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - } + "input": 14.994, + "output": 75.004 } }, - "anthropic/claude-3-opus": { - "id": "anthropic/claude-3-opus", - "name": "Claude Opus 3", - "family": "claude-opus", + "holo3-35b-a3b": { + "id": "holo3-35b-a3b", + "name": "Holo3-35B-A3B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-02-29", - "last_updated": "2024-02-29", + "structured_output": true, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 65536, + "input": 65536, + "output": 65536 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.25, + "output": 1.8 } }, - "anthropic/claude-3.5-haiku": { - "id": "anthropic/claude-3.5-haiku", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", + "qwen3.5-omni-plus": { + "id": "qwen3.5-omni-plus", + "name": "Qwen3.5 Omni Plus", + "description": "Omni-modal model for text, vision, audio, and multimodal agent tasks", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-30", + "last_updated": "2026-03-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 983616, + "input": 983616, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0, + "output": 0 } }, - "anthropic/claude-opus-4": { - "id": "anthropic/claude-opus-4", - "name": "Claude Opus 4", - "family": "claude-opus", + "Gemma-4-31B-it": { + "id": "Gemma-4-31B-it", + "name": "Gemma 4 31B IT", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-09", + "last_updated": "2026-04-09", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.306, + "output": 0.306 } }, - "anthropic/claude-3-haiku": { - "id": "anthropic/claude-3-haiku", - "name": "Claude Haiku 3", - "family": "claude-haiku", + "Qwen3.5-27B-Derestricted": { + "id": "Qwen3.5-27B-Derestricted", + "name": "Qwen3.5 27B Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-13", - "last_updated": "2024-03-13", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 0.306, + "output": 0.306 } }, - "anthropic/claude-sonnet-4.5": { - "id": "anthropic/claude-sonnet-4.5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "sonar-reasoning-pro": { + "id": "sonar-reasoning-pro", + "name": "Perplexity Reasoning Pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 127000, + "input": 127000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 2.006, + "output": 7.9985 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "venice-uncensored:web": { + "id": "venice-uncensored:web", + "name": "Venice Uncensored Web", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-01", + "last_updated": "2024-05-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 80000, + "input": 80000, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.4, + "output": 0.4 } }, - "anthropic/claude-3.5-sonnet": { - "id": "anthropic/claude-3.5-sonnet", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", - "attachment": true, + "doubao-1.5-pro-32k": { + "id": "doubao-1.5-pro-32k", + "name": "Doubao 1.5 Pro 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-22", + "last_updated": "2025-01-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 32000, + "input": 32000, "output": 8192 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.1343, + "output": 0.3349 } }, - "anthropic/claude-3.7-sonnet": { - "id": "anthropic/claude-3.7-sonnet", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "qwen3-30b-a3b-instruct-2507": { + "id": "qwen3-30b-a3b-instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-20", + "last_updated": "2025-02-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.2, + "output": 0.5 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Claude Opus 4", - "family": "claude-opus", + "ernie-5.1:thinking": { + "id": "ernie-5.1:thinking", + "name": "ERNIE 5.1 Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-10", + "last_updated": "2026-05-10", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 119000, + "input": 119000, + "output": 64000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.75, + "output": 3, + "cache_read": 0.75 } }, - "xiaomi/mimo-v2.5-pro": { - "id": "xiaomi/mimo-v2.5-pro", - "name": "MiMo V2.5 Pro", - "family": "mimo-v2.5-pro", - "attachment": true, + "mirothinker-1-7-deepresearch": { + "id": "mirothinker-1-7-deepresearch", + "name": "MiroThinker 1.7 Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-05-01", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 131000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.19999999999999998 + "input": 4, + "output": 25 } }, - "xiaomi/mimo-v2.5": { - "id": "xiaomi/mimo-v2.5", - "name": "MiMo M2.5", - "family": "mimo-v2.5", + "hermes-high": { + "id": "hermes-high", + "name": "Hermes High", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-04-22", - "last_updated": "2026-05-01", + "structured_output": true, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "output": 131100 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.39999999999999997, - "output": 2, - "cache_read": 0.08 + "input": 4.998, + "output": 25.007 } }, - "xiaomi/mimo-v2-pro": { - "id": "xiaomi/mimo-v2-pro", - "name": "MiMo V2 Pro", - "family": "mimo", + "jamba-large-1.6": { + "id": "jamba-large-1.6", + "name": "Jamba Large 1.6", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-20", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 4096 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.19999999999999998 + "input": 1.989, + "output": 7.99 } }, - "xiaomi/mimo-v2-flash": { - "id": "xiaomi/mimo-v2-flash", - "name": "MiMo V2 Flash", - "family": "mimo", + "glm-4-long": { + "id": "glm-4-long", + "name": "GLM-4 Long", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-08-01", + "last_updated": "2024-08-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32000 + "context": 1000000, + "input": 1000000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.29 + "input": 0.2006, + "output": 0.2006 } }, - "bytedance/seed-1.6": { - "id": "bytedance/seed-1.6", - "name": "Seed 1.6", - "family": "seed", - "attachment": false, + "claude-opus-4-thinking:32768": { + "id": "claude-opus-4-thinking:32768", + "name": "Claude 4 Opus Thinking (32K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09", - "last_updated": "2025-09", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, + "context": 200000, + "input": 200000, "output": 32000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.05 + "input": 14.994, + "output": 75.004 } }, - "bytedance/seed-1.8": { - "id": "bytedance/seed-1.8", - "name": "Seed 1.8", - "family": "seed", - "attachment": false, + "claude-sonnet-4-thinking:8192": { + "id": "claude-sonnet-4-thinking:8192", + "name": "Claude 4 Sonnet Thinking (8K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-10", - "last_updated": "2025-10", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, + "context": 1000000, + "input": 1000000, "output": 64000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.05 + "input": 2.992, + "output": 14.994 } }, - "meituan/longcat-flash-chat": { - "id": "meituan/longcat-flash-chat", - "name": "LongCat Flash Chat", - "family": "longcat", + "azure-o3-mini": { + "id": "azure-o3-mini", + "name": "Azure o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-08-30", - "last_updated": "2025-08-30", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 65536 + }, + "cost": { + "input": 1.088, + "output": 4.3996 } }, - "meituan/longcat-flash-thinking": { - "id": "meituan/longcat-flash-thinking", - "name": "LongCat Flash Thinking", - "family": "longcat", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "qvq-max": { + "id": "qvq-max", + "name": "Qwen: QvQ Max", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-28", + "last_updated": "2025-03-28", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 8192 }, "cost": { - "input": 0.15, - "output": 1.5 + "input": 1.4, + "output": 5.3 } }, - "meituan/longcat-flash-thinking-2601": { - "id": "meituan/longcat-flash-thinking-2601", - "name": "LongCat Flash Thinking 2601", - "family": "longcat", - "attachment": false, + "Qwen3.5-27B-BlueStar-v2-Derestricted": { + "id": "Qwen3.5-27B-BlueStar-v2-Derestricted", + "name": "Qwen3.5 27B BlueStar v2 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "temperature": true, - "release_date": "2026-03-13", - "last_updated": "2026-03-13", + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 - } - }, - "bfl/flux-pro-1.0-fill": { - "id": "bfl/flux-pro-1.0-fill", - "name": "FLUX.1 Fill [pro]", - "family": "flux", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2024-10", - "last_updated": "2024-10", - "modalities": { - "input": ["text"], - "output": ["image"] + "context": 262144, + "input": 262144, + "output": 16384 }, - "open_weights": false, - "limit": { - "context": 512, - "output": 0 + "cost": { + "input": 0.306, + "output": 0.306 } }, - "bfl/flux-pro-1.1": { - "id": "bfl/flux-pro-1.1", - "name": "FLUX1.1 [pro]", - "family": "flux", - "attachment": false, - "reasoning": false, + "Qwen3.5-27B-Vivid-Durian": { + "id": "Qwen3.5-27B-Vivid-Durian", + "name": "Qwen3.5 27B Vivid Durian", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "temperature": false, - "release_date": "2024-10", - "last_updated": "2024-10", + "structured_output": false, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text"], - "output": ["image"] + "input": ["text", "image", "video"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 512, - "output": 0 + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 0.306, + "output": 0.306 } }, - "bfl/flux-kontext-pro": { - "id": "bfl/flux-kontext-pro", - "name": "FLUX.1 Kontext Pro", - "family": "flux", + "jamba-large-1.7": { + "id": "jamba-large-1.7", + "name": "Jamba Large 1.7", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-06", - "last_updated": "2025-06", + "structured_output": false, + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 512, - "output": 0 + "context": 256000, + "input": 256000, + "output": 4096 + }, + "cost": { + "input": 1.989, + "output": 7.99 } }, - "bfl/flux-kontext-max": { - "id": "bfl/flux-kontext-max", - "name": "FLUX.1 Kontext Max", - "family": "flux", + "Baichuan-M2": { + "id": "Baichuan-M2", + "name": "Baichuan M2 32B Medical", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-06", - "last_updated": "2025-06", + "structured_output": false, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 512, - "output": 0 + "context": 32768, + "input": 32768, + "output": 32768 + }, + "cost": { + "input": 15.73, + "output": 15.73 } }, - "bfl/flux-pro-1.1-ultra": { - "id": "bfl/flux-pro-1.1-ultra", - "name": "FLUX1.1 [pro] Ultra", - "family": "flux", + "Magistral-Small-2506": { + "id": "Magistral-Small-2506", + "name": "Magistral Small 2506", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2024-11", - "last_updated": "2024-11", + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 512, - "output": 0 + "context": 32768, + "input": 32768, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 1.4 } - } - } - }, - "minimax": { - "id": "minimax", - "env": ["MINIMAX_API_KEY"], - "npm": "@ai-sdk/anthropic", - "api": "https://api.minimax.io/anthropic/v1", - "name": "MiniMax (minimax.io)", - "doc": "https://platform.minimax.io/docs/guides/quickstart", - "models": { - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax-M2", - "family": "minimax", + }, + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek R1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.4, + "output": 1.7 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, + "Qwen3.5-27B-earica-Derestricted-Lite": { + "id": "Qwen3.5-27B-earica-Derestricted-Lite", + "name": "Qwen3.5 27B earica Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 0.306, + "output": 0.306 } }, - "MiniMax-M2.7": { - "id": "MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", + "doubao-seed-2-0-pro-260215": { + "id": "doubao-seed-2-0-pro-260215", + "name": "Doubao Seed 2.0 Pro", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "input": 256000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.782, + "output": 3.876 } }, - "MiniMax-M2.7-highspeed": { - "id": "MiniMax-M2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", - "family": "minimax", + "step-2-mini": { + "id": "step-2-mini", + "name": "Step-2 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-05", + "last_updated": "2024-07-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 8000, + "input": 8000, + "output": 4096 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.2006, + "output": 0.408 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", - "name": "MiniMax-M2.1", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "gemini-exp-1206": { + "id": "gemini-exp-1206", + "name": "Gemini 2.0 Pro 1206", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 2097152, + "input": 2097152, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 1.258, + "output": 4.998 } }, - "MiniMax-M2.5-highspeed": { - "id": "MiniMax-M2.5-highspeed", - "name": "MiniMax-M2.5-highspeed", - "family": "minimax", - "attachment": false, + "claude-opus-4-5-20251101:thinking": { + "id": "claude-opus-4-5-20251101:thinking", + "name": "Claude 4.5 Opus Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "structured_output": true, + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 4.998, + "output": 25.007 } - } - } - }, - "llmgateway": { - "id": "llmgateway", - "env": ["LLMGATEWAY_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.llmgateway.io/v1", - "name": "LLM Gateway", - "doc": "https://llmgateway.io/docs", - "models": { - "gpt-4o-mini-search-preview": { - "id": "gpt-4o-mini-search-preview", - "name": "GPT-4o Mini Search Preview", - "family": "gpt", + }, + "Qwen3.5-27B-BlueStar-v3-Derestricted": { + "id": "Qwen3.5-27B-BlueStar-v3-Derestricted", + "name": "Qwen3.5 27B BlueStar v3 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 262144, + "input": 262144, "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.306, + "output": 0.306 } }, - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "Grok 4.1 Fast Reasoning", - "family": "grok", + "claude-opus-4-thinking": { + "id": "claude-opus-4-thinking", + "name": "Claude 4 Opus Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "release_date": "2025-07-15", + "last_updated": "2025-07-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 14.994, + "output": 75.004 } }, - "qwen3-235b-a22b-instruct-2507": { - "id": "qwen3-235b-a22b-instruct-2507", - "name": "Qwen3 235B A22B Instruct (2507)", - "family": "qwen", - "attachment": false, + "qwen3.5-flash": { + "id": "qwen3.5-flash", + "name": "Qwen3.5 Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 991808, + "input": 991808, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 2.4 + "input": 0.09, + "output": 0.36 } }, - "llama-4-scout": { - "id": "llama-4-scout", - "name": "Llama 4 Scout", - "family": "llama", + "exa-research-pro": { + "id": "exa-research-pro", + "name": "Exa (Research Pro)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-06-04", + "last_updated": "2025-06-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, + "context": 16384, + "input": 16384, "output": 16384 }, - "status": "beta", "cost": { - "input": 0.18, - "output": 0.59 + "input": 2.5, + "output": 2.5 } }, - "hermes-2-pro-llama-3-8b": { - "id": "hermes-2-pro-llama-3-8b", - "name": "Hermes 2 Pro Llama 3 8B", - "family": "hermes", + "jamba-mini": { + "id": "jamba-mini", + "name": "Jamba Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-05-27", - "last_updated": "2024-05-27", + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 4096 }, "cost": { - "input": 0.14, - "output": 0.14 + "input": 0.1989, + "output": 0.408 } }, - "qwen-coder-plus": { - "id": "qwen-coder-plus", - "name": "Qwen Coder Plus", - "family": "qwen", - "attachment": false, - "reasoning": false, + "claude-opus-4-1-thinking:32000": { + "id": "claude-opus-4-1-thinking:32000", + "name": "Claude 4.1 Opus Thinking (32K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.5, - "output": 1 + "input": 14.994, + "output": 75.004 } }, - "auto": { - "id": "auto", - "name": "Auto Route", - "family": "auto", - "attachment": true, + "KAT-Coder-Air-V1": { + "id": "KAT-Coder-Air-V1", + "name": "KAT Coder Air V1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2024-01-01", + "tool_call": false, + "structured_output": false, + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "input": 128000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.2 } }, - "glm-4.6v-flashx": { - "id": "glm-4.6v-flashx", - "name": "GLM-4.6V FlashX", - "family": "glm", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "exa-research": { + "id": "exa-research", + "name": "Exa (Research)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-04", + "last_updated": "2025-06-04", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16000 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 0.04, - "output": 0.4, - "cache_read": 0 + "input": 2.5, + "output": 2.5 } }, - "gemma-2-27b-it-together": { - "id": "gemma-2-27b-it-together", - "name": "Gemma 2 27B IT", - "family": "gemma", - "attachment": false, + "deepclaude": { + "id": "deepclaude", + "name": "DeepClaude", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-06-27", - "last_updated": "2024-06-27", + "release_date": "2025-02-01", + "last_updated": "2025-02-01", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 16384 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.08, - "output": 0.08 + "input": 3, + "output": 15 } }, - "codestral-2508": { - "id": "codestral-2508", - "name": "Codestral", - "family": "mistral", - "attachment": false, - "reasoning": false, - "tool_call": false, + "gemini-2.5-flash-preview-09-2025": { + "id": "gemini-2.5-flash-preview-09-2025", + "name": "Gemini 2.5 Flash Preview (09/2025)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { "input": 0.3, - "output": 0.9 + "output": 2.5 } }, - "gemma-3-1b-it": { - "id": "gemma-3-1b-it", - "name": "Gemma 3 1B IT", - "family": "gemma", + "claude-opus-4-5-20251101": { + "id": "claude-opus-4-5-20251101", + "name": "Claude 4.5 Opus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2025-11-01", + "last_updated": "2025-11-01", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "input": 200000, + "output": 32000 + }, + "cost": { + "input": 4.998, + "output": 25.007 + } + }, + "GLM-4.6-Derestricted-v5": { + "id": "GLM-4.6-Derestricted-v5", + "name": "GLM 4.6 Derestricted v5", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 16384 + "context": 131072, + "input": 131072, + "output": 8192 }, "cost": { - "input": 0.08, - "output": 0.3 + "input": 0.4, + "output": 1.5 } }, - "glm-4-32b-0414-128k": { - "id": "glm-4-32b-0414-128k", - "name": "GLM-4 32B (0414-128k)", - "family": "glm", + "glm-z1-airx": { + "id": "glm-z1-airx", + "name": "GLM Z1 AirX", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 32000, + "input": 32000, "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.7, + "output": 0.7 } }, - "seed-1-6-flash-250715": { - "id": "seed-1-6-flash-250715", - "name": "Seed 1.6 Flash (250715)", - "family": "seed", + "claude-sonnet-4-thinking:1024": { + "id": "claude-sonnet-4-thinking:1024", + "name": "Claude 4 Sonnet Thinking (1K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-07-26", - "last_updated": "2025-07-26", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 0.07, - "output": 0.3, - "cache_read": 0.01 + "input": 2.992, + "output": 14.994 } }, - "seed-1-6-250615": { - "id": "seed-1-6-250615", - "name": "Seed 1.6 (250615)", - "family": "seed", + "holo3-35b-a3b:thinking": { + "id": "holo3-35b-a3b:thinking", + "name": "Holo3-35B-A3B Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-06-25", - "last_updated": "2025-06-25", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 65536, + "input": 65536, + "output": 65536 }, "cost": { "input": 0.25, - "output": 2, - "cache_read": 0.05 + "output": 1.8 } }, - "qwen3-vl-235b-a22b-thinking": { - "id": "qwen3-vl-235b-a22b-thinking", - "name": "Qwen3 VL 235B A22B Thinking", - "family": "qwen", - "attachment": true, - "reasoning": true, + "owl": { + "id": "owl", + "name": "OWL", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1048756, + "input": 1048756, + "output": 262144 }, "cost": { - "input": 0.8, - "output": 2.4 + "input": 0.1, + "output": 0.3 } }, - "qwen3-vl-30b-a3b-thinking": { - "id": "qwen3-vl-30b-a3b-thinking", - "name": "Qwen3 VL 30B A3B Thinking", - "family": "qwen", + "gemini-2.0-pro-reasoner": { + "id": "gemini-2.0-pro-reasoner", + "name": "Gemini 2.0 Pro Reasoner", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-05", + "last_updated": "2025-02-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 128000, + "output": 65536 + }, + "cost": { + "input": 1.292, + "output": 4.998 + } + }, + "Gemma-4-31B-Cognitive-Unshackled": { + "id": "Gemma-4-31B-Cognitive-Unshackled", + "name": "Gemma 4 31B Cognitive Unshackled", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-02", - "last_updated": "2025-10-02", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.306, + "output": 0.306 } }, - "qwen2-5-vl-32b-instruct": { - "id": "qwen2-5-vl-32b-instruct", - "name": "Qwen2.5 VL 32B Instruct", - "family": "qwen", + "Qwen3.5-27B-earica-Derestricted": { + "id": "Qwen3.5-27B-earica-Derestricted", + "name": "Qwen3.5 27B earica Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-03-15", - "last_updated": "2025-03-15", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 0.3 + "input": 0.306, + "output": 0.306 } }, - "qwen3-vl-8b-instruct": { - "id": "qwen3-vl-8b-instruct", - "name": "Qwen3 VL 8B Instruct", - "family": "qwen", + "command-a-plus-05-2026": { + "id": "command-a-plus-05-2026", + "name": "Cohere Command A+ (05/2026)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2025-08-19", - "last_updated": "2025-08-19", + "structured_output": true, + "release_date": "2026-05-22", + "last_updated": "2026-05-22", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 64000 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 2.5, + "output": 10 } }, - "claude-3-7-sonnet": { - "id": "claude-3-7-sonnet", - "name": "Claude 3.7 Sonnet", - "family": "claude", + "auto-model-premium": { + "id": "auto-model-premium", + "name": "Auto model (Premium)", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-02-24", - "last_updated": "2025-02-24", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 1000000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3 + "input": 9.996, + "output": 19.992 } }, - "gemini-pro-latest": { - "id": "gemini-pro-latest", - "name": "Gemini Pro Latest", - "family": "gemini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-27", - "last_updated": "2026-02-27", + "learnlm-1.5-pro-experimental": { + "id": "learnlm-1.5-pro-experimental", + "name": "Gemini LearnLM Experimental", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-14", + "last_updated": "2024-05-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 32767, + "input": 32767, + "output": 8192 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2 + "input": 3.502, + "output": 10.506 } }, - "claude-3-5-haiku": { - "id": "claude-3-5-haiku", - "name": "Claude 3.5 Haiku", - "family": "claude", + "deepseek-r1-sambanova": { + "id": "deepseek-r1-sambanova", + "name": "DeepSeek R1 Fast", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2025-02-20", + "last_updated": "2025-02-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 4096 }, - "status": "deprecated", "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08 + "input": 4.998, + "output": 6.987 } }, - "qwen-max-latest": { - "id": "qwen-max-latest", - "name": "Qwen Max Latest", - "family": "qwen", - "attachment": true, - "reasoning": false, + "claw-medium": { + "id": "claw-medium", + "name": "Claw Medium", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 204800, + "input": 204800, + "output": 131072 }, "cost": { - "input": 1.6, - "output": 6.4 + "input": 0.3, + "output": 1.2 } }, - "glm-4.6v-flash": { - "id": "glm-4.6v-flash", - "name": "GLM-4.6V Flash", - "family": "glm", + "claude-opus-4-20250514": { + "id": "claude-opus-4-20250514", + "name": "Claude 4 Opus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "release_date": "2025-05-14", + "last_updated": "2025-05-14", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16000 + "context": 200000, + "input": 200000, + "output": 32000 }, - "status": "beta", "cost": { - "input": 0, - "output": 0 + "input": 14.994, + "output": 75.004 } }, - "qwen3-30b-a3b-instruct-2507": { - "id": "qwen3-30b-a3b-instruct-2507", - "name": "Qwen3 30B A3B Instruct (2507)", - "family": "qwen", + "yi-large": { + "id": "yi-large", + "name": "Yi Large", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 32000, + "input": 32000, + "output": 4096 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 3.196, + "output": 3.196 } }, - "minimax-text-01": { - "id": "minimax-text-01", - "name": "MiniMax Text 01", - "family": "minimax", + "qwen3-max-2026-01-23": { + "id": "qwen3-max-2026-01-23", + "name": "Qwen3 Max 2026-01-23", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-01-15", - "last_updated": "2025-01-15", + "release_date": "2026-01-26", + "last_updated": "2026-01-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 131072 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 1.1 + "input": 1.2002, + "output": 6.001 } }, - "qwen3-32b-fp8": { - "id": "qwen3-32b-fp8", - "name": "Qwen3 32B FP8", - "family": "qwen", + "phi-4-mini-instruct": { + "id": "phi-4-mini-instruct", + "name": "Phi 4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.17, + "output": 0.68 } }, - "llama-4-scout-17b-instruct": { - "id": "llama-4-scout-17b-instruct", - "name": "Llama 4 Scout 17B Instruct", - "family": "llama", + "ernie-x1-turbo-32k": { + "id": "ernie-x1-turbo-32k", + "name": "Ernie X1 Turbo 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 32000, + "input": 32000, + "output": 16384 }, "cost": { - "input": 0.17, + "input": 0.165, "output": 0.66 } }, - "qwen3-4b-fp8": { - "id": "qwen3-4b-fp8", - "name": "Qwen3 4B FP8", - "family": "qwen", - "attachment": false, + "claw-low": { + "id": "claw-low", + "name": "Claw Low", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1048576, + "input": 1048576, + "output": 65536 }, "cost": { - "input": 0.03, - "output": 0.05 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 } }, - "ministral-8b-2512": { - "id": "ministral-8b-2512", - "name": "Ministral 8B", - "family": "mistral", + "gemini-3-pro-image-preview": { + "id": "gemini-3-pro-image-preview", + "name": "Gemini 3 Pro Image", + "description": "Image model for prompt-driven generation, editing, and visual design workflows", "attachment": true, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "structured_output": false, + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 8192 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 2, + "output": 12 } }, - "gemma-3-27b": { - "id": "gemma-3-27b", - "name": "Gemma 3 27B", - "family": "gemma", + "gemma-4-31B-Garnet": { + "id": "gemma-4-31B-Garnet", + "name": "Gemma 4 31B Garnet", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 262144, + "input": 262144, "output": 16384 }, "cost": { - "input": 0.27, - "output": 0.27 + "input": 0.306, + "output": 0.306 } }, - "qwen3-vl-flash": { - "id": "qwen3-vl-flash", - "name": "Qwen3 VL Flash", - "family": "qwen", + "doubao-1.5-vision-pro-32k": { + "id": "doubao-1.5-vision-pro-32k", + "name": "Doubao 1.5 Vision Pro 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-09", - "last_updated": "2025-10-09", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-22", + "last_updated": "2025-01-22", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32000 + "context": 32000, + "input": 32000, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.01 + "input": 0.459, + "output": 1.377 } }, - "llama-3.1-70b-instruct": { - "id": "llama-3.1-70b-instruct", - "name": "Llama 3.1 70B Instruct", - "family": "llama", + "auto-model-standard": { + "id": "auto-model-standard", + "name": "Auto model (Standard)", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 2048 + "context": 1000000, + "input": 1000000, + "output": 1000000 }, - "status": "beta", "cost": { - "input": 0.72, - "output": 0.72 + "input": 9.996, + "output": 19.992 } }, - "seed-1-8-251228": { - "id": "seed-1-8-251228", - "name": "Seed 1.8 (251228)", - "family": "seed", + "Qwen3.5-27B-Marvin-DPO-V2-Derestricted-Lite": { + "id": "Qwen3.5-27B-Marvin-DPO-V2-Derestricted-Lite", + "name": "Qwen3.5 27B Marvin DPO V2 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-18", - "last_updated": "2025-12-18", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.05 + "input": 0.306, + "output": 0.306 } }, - "qwen3-235b-a22b-thinking-2507": { - "id": "qwen3-235b-a22b-thinking-2507", - "name": "Qwen3 235B A22B Thinking (2507)", - "family": "qwen", + "glm-4": { + "id": "glm-4", + "name": "GLM-4", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-16", + "last_updated": "2024-01-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.8, - "output": 2.4 + "input": 14.994, + "output": 14.994 } }, - "seed-1-6-250915": { - "id": "seed-1-6-250915", - "name": "Seed 1.6 (250915)", - "family": "seed", + "Qwen3.5-27B-Writer-Derestricted-Lite": { + "id": "Qwen3.5-27B-Writer-Derestricted-Lite", + "name": "Qwen3.5 27B Writer Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-09-15", - "last_updated": "2025-09-15", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 256000, - "output": 8192 - }, - "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.05 - } - }, - "glm-4.5-x": { - "id": "glm-4.5-x", - "name": "GLM-4.5 X", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 262144, + "input": 262144, "output": 16384 }, - "status": "beta", "cost": { - "input": 2.2, - "output": 8.9, - "cache_read": 0.45 + "input": 0.306, + "output": 0.306 } }, - "qwen3-30b-a3b-thinking-2507": { - "id": "qwen3-30b-a3b-thinking-2507", - "name": "Qwen3 30B A3B Thinking (2507)", - "family": "qwen", + "qwen-3.6-plus": { + "id": "qwen-3.6-plus", + "name": "Qwen 3.6 Plus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "qwen3.6", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-08", - "last_updated": "2025-07-08", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 991800, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.45, + "output": 2.7 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "Grok 4 Fast Reasoning", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "brave-pro": { + "id": "brave-pro", + "name": "Brave (Pro)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2023-03-02", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 5, + "output": 5 } }, - "deepseek-v3.1": { - "id": "deepseek-v3.1", - "name": "DeepSeek V3.1", - "family": "deepseek", + "deepseek-chat-cheaper": { + "id": "deepseek-chat-cheaper", + "name": "DeepSeek V3/Chat Cheaper", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 32768 + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.56, - "output": 1.68, - "cache_read": 0.11 + "input": 0.25, + "output": 0.7 } }, - "ministral-3b-2512": { - "id": "ministral-3b-2512", - "name": "Ministral 3B", - "family": "mistral", + "Gemma-4-31B-Queen": { + "id": "Gemma-4-31B-Queen", + "name": "Gemma 4 31B Queen", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.306, + "output": 0.306 } }, - "qwen-plus-latest": { - "id": "qwen-plus-latest", - "name": "Qwen Plus Latest", - "family": "qwen", + "Gemma-4-31B-Gemopus": { + "id": "Gemma-4-31B-Gemopus", + "name": "Gemma 4 31B Gemopus", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-01-25", - "last_updated": "2025-01-25", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.306, + "output": 0.306 } }, - "llama-3.1-nemotron-ultra-253b": { - "id": "llama-3.1-nemotron-ultra-253b", - "name": "Llama 3.1 Nemotron Ultra 253B", - "family": "llama", + "mistral-code-latest": { + "id": "mistral-code-latest", + "name": "Mistral Code Latest", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": false, + "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-07", - "last_updated": "2025-04-07", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 0.3, + "output": 0.9 } }, - "llama-4-maverick-17b-instruct": { - "id": "llama-4-maverick-17b-instruct", - "name": "Llama 4 Maverick 17B Instruct", - "family": "llama", + "gemini-2.5-flash-lite": { + "id": "gemini-2.5-flash-lite", + "name": "Gemini 2.5 Flash Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.24, - "output": 0.97 + "input": 0.1, + "output": 0.4 } }, - "grok-4-0709": { - "id": "grok-4-0709", - "name": "Grok 4 (0709)", - "family": "grok", + "jamba-mini-1.7": { + "id": "jamba-mini-1.7", + "name": "Jamba Mini 1.7", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, + "tool_call": false, + "structured_output": false, "release_date": "2025-07-09", "last_updated": "2025-07-09", "modalities": { @@ -99907,6656 +153757,6843 @@ "open_weights": false, "limit": { "context": 256000, - "output": 256000 + "input": 256000, + "output": 4096 }, "cost": { - "input": 3, - "output": 15 + "input": 0.1989, + "output": 0.408 } }, - "qwen3-30b-a3b-fp8": { - "id": "qwen3-30b-a3b-fp8", - "name": "Qwen3 30B A3B FP8", - "family": "qwen", + "universal-summarizer": { + "id": "universal-summarizer", + "name": "Universal Summarizer", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2023-05-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 30, + "output": 30 } }, - "minimax-m2.1-lightning": { - "id": "minimax-m2.1-lightning", - "name": "MiniMax M2.1 Lightning", - "family": "minimax", - "attachment": false, + "Qwen3.5-27B-Anko": { + "id": "Qwen3.5-27B-Anko", + "name": "Qwen3.5 27B Anko", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 131072 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.12, - "output": 0.48 + "input": 0.306, + "output": 0.306 } }, - "qwen3-max-2026-01-23": { - "id": "qwen3-max-2026-01-23", - "name": "Qwen3 Max (2026-01-23)", - "family": "qwen", + "claude-sonnet-4-20250514": { + "id": "claude-sonnet-4-20250514", + "name": "Claude 4 Sonnet", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-01-23", - "last_updated": "2026-01-23", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 32800 + "context": 200000, + "input": 200000, + "output": 64000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.6 + "input": 2.992, + "output": 14.994 } }, - "llama-3.2-3b-instruct": { - "id": "llama-3.2-3b-instruct", - "name": "Llama 3.2 3B Instruct", - "family": "llama", + "ernie-x1-32k-preview": { + "id": "ernie-x1-32k-preview", + "name": "Ernie X1 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-18", - "last_updated": "2024-09-18", + "structured_output": false, + "release_date": "2025-04-03", + "last_updated": "2025-04-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 32000 + "context": 32000, + "input": 32000, + "output": 16384 }, "cost": { - "input": 0.03, - "output": 0.05 + "input": 0.33, + "output": 1.32 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "Qwen3 Coder Next", - "family": "qwen", + "azure-gpt-4-turbo": { + "id": "azure-gpt-4-turbo", + "name": "Azure gpt-4-turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2023-11-06", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.8, - "output": 4 + "input": 9.996, + "output": 30.005 } }, - "gpt-4o-search-preview": { - "id": "gpt-4o-search-preview", - "name": "GPT-4o Search Preview", - "family": "gpt", - "attachment": true, + "Meta-Llama-3-1-8B-Instruct-FP8": { + "id": "Meta-Llama-3-1-8B-Instruct-FP8", + "name": "Llama 3.1 8B (decentralized)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 16384 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.02, + "output": 0.03 } }, - "custom": { - "id": "custom", - "name": "Custom Model", - "family": "auto", + "claude-sonnet-4-5-20250929-thinking": { + "id": "claude-sonnet-4-5-20250929-thinking", + "name": "Claude Sonnet 4.5 Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2024-01-01", - "last_updated": "2024-01-01", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 1000000, + "output": 64000 + }, + "cost": { + "input": 2.992, + "output": 14.994 + } + }, + "asi1-mini": { + "id": "asi1-mini", + "name": "ASI1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-25", + "last_updated": "2025-03-25", + "modalities": { + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 1 } }, - "qwen3-vl-30b-a3b-instruct": { - "id": "qwen3-vl-30b-a3b-instruct", - "name": "Qwen3 VL 30B A3B Instruct", - "family": "qwen", + "qwen3.5-27b": { + "id": "qwen3.5-27b", + "name": "Qwen3.5 27B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-10-02", - "last_updated": "2025-10-02", + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 260096, + "input": 260096, + "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0.27, + "output": 2.16 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek V3.2", - "family": "deepseek", + "Qwen3.5-27B-Omega-Evolution-v2.2-Derestricted-Lite": { + "id": "Qwen3.5-27B-Omega-Evolution-v2.2-Derestricted-Lite", + "name": "Qwen3.5 27B Omega Evolution v2.2 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, + "context": 262144, + "input": 262144, "output": 16384 }, "cost": { - "input": 0.28, - "output": 0.42, - "cache_read": 0.03 + "input": 0.306, + "output": 0.306 } }, - "qwen3-235b-a22b-fp8": { - "id": "qwen3-235b-a22b-fp8", - "name": "Qwen3 235B A22B FP8", - "family": "qwen", - "attachment": false, + "claude-opus-4-1-thinking:32768": { + "id": "claude-opus-4-1-thinking:32768", + "name": "Claude 4.1 Opus Thinking (32K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.5, - "output": 2.5 + "input": 14.994, + "output": 75.004 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, + "gemma-4-31B-K1-v5": { + "id": "gemma-4-31B-K1-v5", + "name": "Gemma 4 31B K1 v5", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32766 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.5 + "input": 0.306, + "output": 0.306 } }, - "kimi-k2": { - "id": "kimi-k2", - "name": "Kimi K2", - "family": "kimi", + "sonar": { + "id": "sonar", + "name": "Perplexity Simple", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 127000, + "input": 127000, + "output": 128000 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.5 + "input": 1.003, + "output": 1.003 } }, - "llama-3-8b-instruct": { - "id": "llama-3-8b-instruct", - "name": "Llama 3 8B Instruct", - "family": "llama", + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax M2", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2025-04-03", - "last_updated": "2025-04-03", + "structured_output": false, + "release_date": "2025-10-25", + "last_updated": "2025-10-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 131072 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 0.17, + "output": 1.53 } }, - "qwen3-vl-235b-a22b-instruct": { - "id": "qwen3-vl-235b-a22b-instruct", - "name": "Qwen3 VL 235B A22B Instruct", - "family": "qwen", - "attachment": true, - "reasoning": false, + "sarvam-105b": { + "id": "sarvam-105b", + "name": "Sarvam 105B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "structured_output": false, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, - "output": 8192 + "input": 131072, + "output": 4096 }, "cost": { - "input": 0.8, - "output": 2.4 + "input": 0.045, + "output": 0.177, + "cache_read": 0.028 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + "Qwen3.5-27B-Writer-V2-Derestricted-Lite": { + "id": "Qwen3.5-27B-Writer-V2-Derestricted-Lite", + "name": "Qwen3.5 27B Writer V2 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32766 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.75 + "input": 0.306, + "output": 0.306 } }, - "qwen25-coder-7b": { - "id": "qwen25-coder-7b", - "name": "Qwen2.5 Coder 7B", - "family": "qwen", - "attachment": false, - "reasoning": false, + "Qwen3.5-27B-Marvin-V2-Derestricted": { + "id": "Qwen3.5-27B-Marvin-V2-Derestricted", + "name": "Qwen3.5 27B Marvin V2 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-19", - "last_updated": "2024-09-19", + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.05 + "input": 0.306, + "output": 0.306 } }, - "llama-3.1-8b-instruct": { - "id": "llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "claw-high": { + "id": "claw-high", + "name": "Claw High", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 2048 + "context": 1000000, + "input": 1000000, + "output": 128000 }, - "status": "beta", "cost": { - "input": 0.22, - "output": 0.22 + "input": 4.998, + "output": 25.007 } }, - "llama-3-70b-instruct": { - "id": "llama-3-70b-instruct", - "name": "Llama 3 70B Instruct", - "family": "llama", + "qwen-max": { + "id": "qwen-max", + "name": "Qwen 2.5 Max", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "structured_output": false, + "release_date": "2024-04-03", + "last_updated": "2024-04-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8000 + "context": 32000, + "input": 32000, + "output": 8192 }, "cost": { - "input": 0.51, - "output": 0.74 + "input": 1.5997, + "output": 6.392 } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek R1 (0528)", - "family": "deepseek", - "attachment": false, + "gemma-4-31B-Fabled": { + "id": "gemma-4-31B-Fabled", + "name": "Gemma 4 31B Fabled", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, + "context": 262144, + "input": 262144, "output": 16384 }, - "status": "beta", "cost": { - "input": 0.8, - "output": 2.4 + "input": 0.306, + "output": 0.306 } }, - "glm-4.5-airx": { - "id": "glm-4.5-airx", - "name": "GLM-4.5 AirX", - "family": "glm", + "v0-1.5-md": { + "id": "v0-1.5-md", + "name": "v0 1.5 MD", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-04", + "last_updated": "2025-07-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "input": 200000, + "output": 64000 }, "cost": { - "input": 1.1, - "output": 4.5, - "cache_read": 0.22 + "input": 3, + "output": 15 } }, - "ministral-14b-2512": { - "id": "ministral-14b-2512", - "name": "Ministral 14B", - "family": "mistral", + "claude-opus-4-thinking:1024": { + "id": "claude-opus-4-thinking:1024", + "name": "Claude 4 Opus Thinking (1K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-02", - "last_updated": "2025-12-02", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.2 + "input": 14.994, + "output": 75.004 } }, - "llama-3.2-11b-instruct": { - "id": "llama-3.2-11b-instruct", - "name": "Llama 3.2 11B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "gemini-2.5-flash-preview-04-17": { + "id": "gemini-2.5-flash-preview-04-17", + "name": "Gemini 2.5 Flash Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": false, - "structured_output": true, - "temperature": true, - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "structured_output": false, + "release_date": "2025-04-17", + "last_updated": "2025-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.33 + "input": 0.15, + "output": 0.6 } }, - "claude-3-opus": { - "id": "claude-3-opus", - "name": "Claude 3 Opus", - "family": "claude", + "gemini-2.5-pro-exp-03-25": { + "id": "gemini-2.5-pro-exp-03-25", + "name": "Gemini 2.5 Pro Experimental 0325", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-03-04", - "last_updated": "2024-03-04", + "release_date": "2025-03-25", + "last_updated": "2025-03-25", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 4096 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5 + "input": 2.5, + "output": 10 } }, - "minimax-m2.7": { - "id": "minimax-m2.7", - "name": "MiniMax-M2.7", - "family": "minimax", + "hermes-medium": { + "id": "hermes-medium", + "name": "Hermes Medium", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "release_date": "2026-05-11", + "last_updated": "2026-05-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 204800, + "input": 204800, "output": 131072 }, "cost": { "input": 0.3, - "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "output": 1.2 } }, - "grok-4-20-beta-0309-non-reasoning": { - "id": "grok-4-20-beta-0309-non-reasoning", - "name": "Grok 4.20 (Non-Reasoning)", - "family": "grok", - "attachment": true, + "sonar-pro": { + "id": "sonar-pro", + "name": "Perplexity Pro", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-09", + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 2.992, + "output": 14.994 } }, - "qwen3-coder-plus": { - "id": "qwen3-coder-plus", - "name": "Qwen3 Coder Plus", - "family": "qwen", - "attachment": false, + "doubao-1-5-thinking-pro-250415": { + "id": "doubao-1-5-thinking-pro-250415", + "name": "Doubao 1.5 Thinking Pro", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-17", + "last_updated": "2025-04-17", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 1, - "output": 5 + "input": 0.6, + "output": 2.4 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5 (latest)", - "family": "claude-haiku", - "attachment": true, + "qwen3.7-max:thinking": { + "id": "qwen3.7-max:thinking", + "name": "Qwen3.7 Max Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 262144 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-21", + "last_updated": "2026-05-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "input": 1000000, + "output": 65536 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 2.5, + "output": 7.5, + "cache_read": 0.25 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "ernie-4.5-turbo-vl-32k": { + "id": "ernie-4.5-turbo-vl-32k", + "name": "Ernie 4.5 Turbo VL 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 32000, + "input": 32000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.495, + "output": 1.43 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview 09-25", - "family": "gemini-flash-lite", + "gemini-2.5-pro-preview-05-06": { + "id": "gemini-2.5-pro-preview-05-06", + "name": "Gemini 2.5 Pro Preview 0506", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-06", + "last_updated": "2025-05-06", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1048756, + "input": 1048756, "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 2.5, + "output": 10 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi-k2.5", + "hunyuan-turbos-20250226": { + "id": "hunyuan-turbos-20250226", + "name": "Hunyuan Turbo S", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 24000, + "input": 24000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.187, + "output": 0.374 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", + "claude-sonnet-4-thinking:64000": { + "id": "claude-sonnet-4-thinking:64000", + "name": "Claude 4 Sonnet Thinking (64K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 2.992, + "output": 14.994 } }, - "mistral-large-2512": { - "id": "mistral-large-2512", - "name": "Mistral Large 3", - "family": "mistral-large", + "ernie-x1-32k": { + "id": "ernie-x1-32k", + "name": "Ernie X1 32k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2025-12-02", + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 32000, + "input": 32000, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.33, + "output": 1.32 } }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", + "command-a-reasoning-08-2025": { + "id": "command-a-reasoning-08-2025", + "name": "Cohere Command A (08/2025)", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-22", + "last_updated": "2025-08-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "input": 256000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 2.5, + "output": 10 } }, - "minimax-m2.7-highspeed": { - "id": "minimax-m2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", - "family": "minimax", + "doubao-seed-2-0-code-preview-260215": { + "id": "doubao-seed-2-0-code-preview-260215", + "name": "Doubao Seed 2.0 Code Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-14", + "last_updated": "2026-02-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 256000, + "input": 256000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.782, + "output": 3.893 } }, - "mimo-v2.5-pro": { - "id": "mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", - "family": "mimo", + "doubao-1.5-pro-256k": { + "id": "doubao-1.5-pro-256k", + "name": "Doubao 1.5 Pro 256k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-12", + "last_updated": "2025-03-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.799, + "output": 1.445 } }, - "gemma-3n-e4b-it": { - "id": "gemma-3n-e4b-it", - "name": "Gemma 3n 4B", - "family": "gemma", + "qwen3.5-35b-a3b": { + "id": "qwen3.5-35b-a3b", + "name": "Qwen3.5 35B A3B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2000 + "context": 260096, + "input": 260096, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.225, + "output": 1.8 } }, - "claude-3-5-sonnet-20241022": { - "id": "claude-3-5-sonnet-20241022", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", + "qwen3.5-122b-a10b:thinking": { + "id": "qwen3.5-122b-a10b:thinking", + "name": "Qwen3.5 122B A10B Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 260096, + "input": 260096, + "output": 65536 + }, + "cost": { + "input": 0.36, + "output": 2.88 + } + }, + "yi-lightning": { + "id": "yi-lightning", + "name": "Yi Lightning", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2024-10-16", + "last_updated": "2024-10-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 12000, + "input": 12000, + "output": 4096 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.2006, + "output": 0.2006 } }, - "gpt-5.2-pro": { - "id": "gpt-5.2-pro", - "name": "GPT-5.2 Pro", - "family": "gpt-pro", + "claude-sonnet-4-5-20250929": { + "id": "claude-sonnet-4-5-20250929", + "name": "Claude Sonnet 4.5", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "structured_output": true, + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1000000, + "input": 1000000, + "output": 64000 }, "cost": { - "input": 21, - "output": 168 + "input": 2.992, + "output": 14.994 } }, - "qwq-plus": { - "id": "qwq-plus", - "name": "QwQ Plus", - "family": "qwen", + "deepseek-math-v2": { + "id": "deepseek-math-v2", + "name": "DeepSeek Math V2", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-03", + "last_updated": "2025-12-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 2.4 + "input": 0.6, + "output": 2.2 } }, - "gemini-3.1-flash-lite-preview": { - "id": "gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", + "claude-opus-4-1-thinking:8192": { + "id": "claude-opus-4-1-thinking:8192", + "name": "Claude 4.1 Opus Thinking (8K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 14.994, + "output": 75.004 } }, - "qwen-vl-plus": { - "id": "qwen-vl-plus", - "name": "Qwen-VL Plus", - "family": "qwen", + "deepseek-reasoner": { + "id": "deepseek-reasoner", + "name": "DeepSeek Reasoner", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-08-15", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 64000, + "input": 64000, + "output": 65536 }, "cost": { - "input": 0.21, - "output": 0.63 + "input": 0.4, + "output": 1.7 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "gemini-2.5-flash-nothinking": { + "id": "gemini-2.5-flash-nothinking", + "name": "Gemini 2.5 Flash (No Thinking)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2, - "cache_write": 0 + "input": 0.3, + "output": 2.5 } }, - "devstral-2512": { - "id": "devstral-2512", - "name": "Devstral 2", - "family": "devstral", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-12", - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "Qwen3.5-27B-Omega-Evolution-v2.0-Derestricted-Lite": { + "id": "Qwen3.5-27B-Omega-Evolution-v2.0-Derestricted-Lite", + "name": "Qwen3.5 27B Omega Evolution v2.0 Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 262144 + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.306, + "output": 0.306 } }, - "qwen3-32b": { - "id": "qwen3-32b", - "name": "Qwen3 32B", - "family": "qwen", + "fastgpt": { + "id": "fastgpt", + "name": "Web Answer", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2023-08-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.7, - "output": 2.8, - "reasoning": 8.4 + "input": 7.5, + "output": 7.5 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", + "Qwen3.5-27B-Infracelestial": { + "id": "Qwen3.5-27B-Infracelestial", + "name": "Qwen3.5 27B Infracelestial", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.306, + "output": 0.306 } }, - "glm-4.7-flashx": { - "id": "glm-4.7-flashx", - "name": "GLM-4.7-FlashX", - "family": "glm-flash", + "glm-4-flash": { + "id": "glm-4-flash", + "name": "GLM-4 Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-08-01", + "last_updated": "2024-08-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.07, - "output": 0.4, - "cache_read": 0.01, - "cache_write": 0 + "input": 0.1003, + "output": 0.1003 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "azure-gpt-4o-mini": { + "id": "azure-gpt-4o-mini", + "name": "Azure gpt-4o-mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.1496, + "output": 0.595 } }, - "qwen35-397b-a17b": { - "id": "qwen35-397b-a17b", - "name": "Qwen3.5 397B-A17B", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-02-15", - "last_updated": "2026-02-15", + "sonar-deep-research": { + "id": "sonar-deep-research", + "name": "Perplexity Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-25", + "last_updated": "2025-02-25", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 60000, + "input": 60000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 3.4, + "output": 13.6 } }, - "qwen-max": { - "id": "qwen-max", - "name": "Qwen Max", - "family": "qwen", + "qwq-32b": { + "id": "qwq-32b", + "name": "Qwen: QwQ 32B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-03", - "last_updated": "2025-01-25", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 32768 }, "cost": { - "input": 1.6, - "output": 6.4 + "input": 0.25599999, + "output": 0.30499999 } }, - "gpt-5.3-chat-latest": { - "id": "gpt-5.3-chat-latest", - "name": "GPT-5.3 Chat (latest)", - "family": "gpt", - "attachment": true, + "mistral-code-agent-latest": { + "id": "mistral-code-agent-latest", + "name": "Mistral Code Agent Latest", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2026-06-02", + "last_updated": "2026-06-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 262144, + "input": 262144, + "output": 32768 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.4, + "output": 2 } }, - "gemini-2.0-flash": { - "id": "gemini-2.0-flash", - "name": "Gemini 2.0 Flash", - "family": "gemini-flash", + "glm-4.1v-thinking-flash": { + "id": "glm-4.1v-thinking-flash", + "name": "GLM 4.1V Thinking Flash", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-09", + "last_updated": "2025-07-09", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 64000, + "input": 64000, "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.3, + "output": 0.3 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", + "qwen3.5-omni-flash": { + "id": "qwen3.5-omni-flash", + "name": "Qwen3.5 Omni Flash", + "description": "Omni-modal model for text, vision, audio, and multimodal agent tasks", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-30", + "last_updated": "2026-03-30", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 49152, + "input": 49152, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "input_audio": 1 + "input": 0, + "output": 0 } }, - "qwen-plus": { - "id": "qwen-plus", - "name": "Qwen Plus", - "family": "qwen", - "attachment": false, + "gemma-4-31B-MeroMero": { + "id": "gemma-4-31B-MeroMero", + "name": "Gemma 4 31B MeroMero", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-01-25", - "last_updated": "2025-09-11", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.4, - "output": 1.2, - "reasoning": 4 + "input": 0.306, + "output": 0.306 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", + "gemini-2.5-flash-preview-04-17:thinking": { + "id": "gemini-2.5-flash-preview-04-17:thinking", + "name": "Gemini 2.5 Flash Preview Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-17", + "last_updated": "2025-04-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 1048756, + "input": 1048756, + "output": 65536 }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 12.5, - "output": 75, - "cache_read": 1.25 - }, - "provider": { - "body": { - "service_tier": "priority" - } - } - } - } + "cost": { + "input": 0.15, + "output": 3.5 + } + }, + "glm-4-air": { + "id": "glm-4-air", + "name": "GLM-4 Air", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-06-05", + "last_updated": "2024-06-05", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.2006, + "output": 0.2006 } }, - "qwen3.6-35b-a3b": { - "id": "qwen3.6-35b-a3b", - "name": "Qwen3.6 35B-A3B", - "family": "qwen", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-17", - "last_updated": "2026-04-17", + "doubao-seed-1-6-thinking-250615": { + "id": "doubao-seed-1-6-thinking-250615", + "name": "Doubao Seed 1.6 Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-15", + "last_updated": "2025-06-15", "modalities": { - "input": ["text", "image", "video", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 0.248, - "output": 1.485 + "input": 0.204, + "output": 2.04 } }, - "qwen-omni-turbo": { - "id": "qwen-omni-turbo", - "name": "Qwen-Omni Turbo", - "family": "qwen", + "auto-model-basic": { + "id": "auto-model-basic", + "name": "Auto model (Basic)", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-01-19", - "last_updated": "2025-03-26", + "tool_call": false, + "structured_output": false, + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 2048 + "context": 1000000, + "input": 1000000, + "output": 1000000 }, "cost": { - "input": 0.07, - "output": 0.27, - "input_audio": 4.44, - "output_audio": 8.89 + "input": 9.996, + "output": 19.992 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "Qwen3.5-27B-RpRMax-v1": { + "id": "Qwen3.5-27B-RpRMax-v1", + "name": "Qwen3.5 27B RpRMax v1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.306, + "output": 0.306 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "ernie-5.1": { + "id": "ernie-5.1", + "name": "ERNIE 5.1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-10", + "last_updated": "2026-05-10", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 119000, + "input": 119000, + "output": 64000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.75, + "output": 3, + "cache_read": 0.75 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", + "Qwen3.5-27B-Marvin-DPO-V2-Derestricted": { + "id": "Qwen3.5-27B-Marvin-DPO-V2-Derestricted", + "name": "Qwen3.5 27B Marvin DPO V2 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 262144, + "input": 262144, + "output": 16384 + }, + "cost": { + "input": 0.306, + "output": 0.306 + } + }, + "azure-gpt-4o": { + "id": "azure-gpt-4o", + "name": "Azure gpt-4o", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 2.499, + "output": 9.996 } }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo-V2-Omni", - "family": "mimo", + "deepseek-chat": { + "id": "deepseek-chat", + "name": "DeepSeek V3/Deepseek Chat", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 0.25, + "output": 0.7 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", + "gemini-2.5-flash-preview-05-20": { + "id": "gemini-2.5-flash-preview-05-20", + "name": "Gemini 2.5 Flash 0520", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048000, + "input": 1048000, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.15, + "output": 0.6 } }, - "minimax-m2": { - "id": "minimax-m2", - "name": "MiniMax-M2", - "family": "minimax", + "mercury-2": { + "id": "mercury-2", + "name": "Mercury 2", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "structured_output": true, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 50000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.25, + "output": 0.75, + "cache_read": 0.025 } - }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", + }, + "qwen3.7-plus:thinking": { + "id": "qwen3.7-plus:thinking", + "name": "Qwen3.7 Plus Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 262144 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 983616, + "input": 983616, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.4, + "output": 1.6, + "cache_read": 0.04 } }, - "qwen-flash": { - "id": "qwen-flash", - "name": "Qwen Flash", - "family": "qwen", + "gemini-2.0-flash-thinking-exp-1219": { + "id": "gemini-2.0-flash-thinking-exp-1219", + "name": "Gemini 2.0 Flash Thinking 1219", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-19", + "last_updated": "2024-12-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 32768 + "context": 32767, + "input": 32767, + "output": 8192 }, "cost": { - "input": 0.05, - "output": 0.4 + "input": 0.1003, + "output": 0.408 } }, - "gpt-4-turbo": { - "id": "gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", - "attachment": true, + "glm-4-plus-0111": { + "id": "glm-4-plus-0111", + "name": "GLM 4 Plus 0111", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2025-02-19", + "last_updated": "2025-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 4096 }, "cost": { - "input": 10, - "output": 30 + "input": 9.996, + "output": 9.996 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "brave": { + "id": "brave", + "name": "Brave (Answers)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2023-03-02", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - }, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 5, + "output": 5 } }, - "mimo-v2.5": { - "id": "mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "glm-zero-preview": { + "id": "glm-zero-preview", + "name": "GLM Zero Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 8000, + "input": 8000, + "output": 4096 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, - "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 - }, - "tiers": [ - { - "input": 0.8, - "output": 4, - "cache_read": 0.16, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 1.802, + "output": 1.802 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", + "gemini-2.5-flash-lite-preview-06-17": { + "id": "gemini-2.5-flash-lite-preview-06-17", + "name": "Gemini 2.5 Flash Lite Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.15, + "output": 0.6 } }, - "sonar-pro": { - "id": "sonar-pro", - "name": "Sonar Pro", - "family": "sonar-pro", + "Qwen3.5-27B-Writer-Derestricted": { + "id": "Qwen3.5-27B-Writer-Derestricted", + "name": "Qwen3.5 27B Writer Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 3, - "output": 15 + "input": 0.306, + "output": 0.306 } }, - "pixtral-large-latest": { - "id": "pixtral-large-latest", - "name": "Pixtral Large (latest)", - "family": "pixtral", - "attachment": true, + "KAT-Coder-Exp-72B-1010": { + "id": "KAT-Coder-Exp-72B-1010", + "name": "KAT Coder Exp 72B 1010", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2024-11-04", + "tool_call": false, + "structured_output": false, + "release_date": "2025-10-28", + "last_updated": "2025-10-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 128000 + "input": 128000, + "output": 32768 }, "cost": { - "input": 2, - "output": 6 + "input": 0.1, + "output": 0.2 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", + "gemini-2.5-pro-preview-06-05": { + "id": "gemini-2.5-pro-preview-06-05", + "name": "Gemini 2.5 Pro Preview 0605", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-05", + "last_updated": "2025-06-05", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 2.5, + "output": 10 } }, - "grok-4-20-beta-0309-reasoning": { - "id": "grok-4-20-beta-0309-reasoning", - "name": "Grok 4.20 (Reasoning)", - "family": "grok", + "Qwen3.5-27B-Musica-v1": { + "id": "Qwen3.5-27B-Musica-v1", + "name": "Qwen3.5 27B Musica v1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-09", - "last_updated": "2026-03-09", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 12, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 12, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.306, + "output": 0.306 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", + "MiniMax-M1": { + "id": "MiniMax-M1", + "name": "MiniMax M1", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-16", + "last_updated": "2025-06-16", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 1000000, + "output": 131072 + }, + "cost": { + "input": 0.1394, + "output": 1.3328 + } + }, + "doubao-1-5-thinking-pro-vision-250415": { + "id": "doubao-1-5-thinking-pro-vision-250415", + "name": "Doubao 1.5 Thinking Pro Vision", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 16384 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 0.6, + "output": 2.4 } }, - "qwen3.6-plus": { - "id": "qwen3.6-plus", - "name": "Qwen3.6 Plus", - "family": "qwen", - "attachment": false, + "qwen3.5-27b:thinking": { + "id": "qwen3.5-27b:thinking", + "name": "Qwen3.5 27B Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 260096, + "input": 260096, "output": 65536 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "cache_write": 0.625, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.2, - "cache_write": 2.5, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.27, + "output": 2.16 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 mini", - "family": "gpt-mini", + "Qwen3.5-27B-Omega-Evolution-v2.0-Derestricted": { + "id": "Qwen3.5-27B-Omega-Evolution-v2.0-Derestricted", + "name": "Qwen3.5 27B Omega Evolution v2.0 Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0.306, + "output": 0.306 } }, - "qwen3-max": { - "id": "qwen3-max", - "name": "Qwen3 Max", - "family": "qwen", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "Gemma-4-31B-GarnetV2": { + "id": "Gemma-4-31B-GarnetV2", + "name": "Gemma 4 31B Garnet V2", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-01", + "last_updated": "2026-05-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { "context": 262144, - "output": 65536 + "input": 262144, + "output": 16384 }, "cost": { - "input": 1.2, - "output": 6 + "input": 0.306, + "output": 0.306 } }, - "minimax-m2.1": { - "id": "minimax-m2.1", - "name": "MiniMax-M2.1", - "family": "minimax", + "qwen-turbo": { + "id": "qwen-turbo", + "name": "Qwen Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-11-01", + "last_updated": "2024-11-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1000000, + "input": 1000000, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.04998, + "output": 0.2006 } }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", + "phi-4-multimodal-instruct": { + "id": "phi-4-multimodal-instruct", + "name": "Phi 4 Multimodal", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 6, - "output": 24, - "cache_read": 1.3, - "cache_write": 0 + "input": 0.07, + "output": 0.11 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", + "mistral-small-31-24b-instruct": { + "id": "mistral-small-31-24b-instruct", + "name": "Mistral Small 31 24b Instruct", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "input": 128000, + "output": 131072 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 0.1, + "output": 0.3 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 nano", - "family": "gpt-nano", + "ernie-4.5-turbo-128k": { + "id": "ernie-4.5-turbo-128k", + "name": "Ernie 4.5 Turbo 128k", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-08", + "last_updated": "2025-05-08", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 0.132, + "output": 0.55 } }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, + "Qwen3.5-27B-BlueStar-Derestricted": { + "id": "Qwen3.5-27B-BlueStar-Derestricted", + "name": "Qwen3.5 27B BlueStar Derestricted", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-06", + "last_updated": "2026-04-06", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 0.306, + "output": 0.306 } }, - "mistral-large-latest": { - "id": "mistral-large-latest", - "name": "Mistral Large (latest)", - "family": "mistral-large", + "gemini-2.5-flash-lite-preview-09-2025-thinking": { + "id": "gemini-2.5-flash-lite-preview-09-2025-thinking", + "name": "Gemini 2.5 Flash Lite Preview (09/2025) – Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2024-11-01", - "last_updated": "2025-12-02", + "structured_output": true, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.1, + "output": 0.4 } }, - "mistral-small-2506": { - "id": "mistral-small-2506", - "name": "Mistral Small 3.2", - "family": "mistral-small", + "doubao-seed-1-8-251215": { + "id": "doubao-seed-1-8-251215", + "name": "Doubao Seed 1.8", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-06-20", - "last_updated": "2025-06-20", + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.612, + "output": 6.12 } }, - "gemma-3-12b-it": { - "id": "gemma-3-12b-it", - "name": "Gemma 3 12B", - "family": "gemma", - "attachment": true, + "qwen3.6-max-preview": { + "id": "qwen3.6-max-preview", + "name": "Qwen3.6 Max Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "qwen3.6", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "structured_output": false, + "release_date": "2026-04-20", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 245800, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.3, + "output": 7.8 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "exa-answer": { + "id": "exa-answer", + "name": "Exa (Answer)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-04", + "last_updated": "2025-06-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 4096, + "input": 4096, + "output": 4096 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 2.5, + "output": 2.5 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "Baichuan4-Air": { + "id": "Baichuan4-Air", + "name": "Baichuan 4 Air", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "input_audio": 1 + "input": 0.157, + "output": 0.157 } }, - "gpt-5.2-chat-latest": { - "id": "gpt-5.2-chat-latest", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", + "qwen3.5-122b-a10b": { + "id": "qwen3.5-122b-a10b", + "name": "Qwen3.5 122B A10B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 260096, + "input": 260096, + "output": 65536 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.36, + "output": 2.88 } }, - "gemma-3n-e2b-it": { - "id": "gemma-3n-e2b-it", - "name": "Gemma 3n 2B", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "sarvam-30b": { + "id": "sarvam-30b", + "name": "Sarvam 30B", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": false, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2000 + "context": 65536, + "input": 65536, + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.028, + "output": 0.111, + "cache_read": 0.017 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex mini", - "family": "gpt-codex", + "claude-opus-4-1-thinking": { + "id": "claude-opus-4-1-thinking", + "name": "Claude 4.1 Opus Thinking", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 14.994, + "output": 75.004 } }, - "grok-4-fast": { - "id": "grok-4-fast", - "name": "Grok 4 Fast", - "family": "grok", + "claude-opus-4-thinking:32000": { + "id": "claude-opus-4-thinking:32000", + "name": "Claude 4 Opus Thinking (32K)", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "structured_output": true, + "release_date": "2025-05-22", + "last_updated": "2025-05-22", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 200000, + "input": 200000, + "output": 32000 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 14.994, + "output": 75.004 } }, - "gemini-3.1-flash-lite": { - "id": "gemini-3.1-flash-lite", - "name": "Gemini 3.1 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-05-07", - "last_updated": "2026-05-07", + "auto-model": { + "id": "auto-model", + "name": "Auto model", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-06-01", + "last_updated": "2024-06-01", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "input": 1000000, + "output": 1000000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 0, + "output": 0 } }, - "qwen3-next-80b-a3b-thinking": { - "id": "qwen3-next-80b-a3b-thinking", - "name": "Qwen3-Next 80B-A3B (Thinking)", - "family": "qwen", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09", - "last_updated": "2025-09", + "qwen3-vl-235b-a22b-instruct-original": { + "id": "qwen3-vl-235b-a22b-instruct-original", + "name": "Qwen3 VL 235B A22B Instruct Original", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 32768, + "input": 32768, "output": 32768 }, "cost": { "input": 0.5, - "output": 6 + "output": 1.2 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", + "glm-z1-air": { + "id": "glm-z1-air", + "name": "GLM Z1 Air", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "structured_output": true, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 32000, + "input": 32000, + "output": 16384 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.07, + "output": 0.07 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "Qwen3.5-27B-Queen-Derestricted-Lite": { + "id": "Qwen3.5-27B-Queen-Derestricted-Lite", + "name": "Qwen3.5 27B Queen Derestricted Lite", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.306, + "output": 0.306 } }, - "gemma-3-4b-it": { - "id": "gemma-3-4b-it", - "name": "Gemma 3 4B", - "family": "gemma", - "attachment": true, + "inclusionai/ling-2.6-1t": { + "id": "inclusionai/ling-2.6-1t", + "name": "Ling 2.6 1T", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "tool_call": true, + "structured_output": true, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 32768 }, - "cost": { - "input": 0, - "output": 0 + "cost": { + "input": 0.3, + "output": 2.5, + "cache_read": 0.06 } }, - "kimi-k2-thinking-turbo": { - "id": "kimi-k2-thinking-turbo", - "name": "Kimi K2 Thinking Turbo", - "family": "kimi-thinking", + "inclusionai/ring-2.6-1t": { + "id": "inclusionai/ring-2.6-1t", + "name": "Ring 2.6 1T", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "structured_output": true, + "release_date": "2026-05-08", + "last_updated": "2026-05-08", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 262144, - "output": 262144 + "input": 262144, + "output": 65536 }, "cost": { - "input": 1.15, - "output": 8, - "cache_read": 0.15 + "input": 1, + "output": 3 } }, - "o1": { - "id": "o1", - "name": "o1", - "family": "o", - "attachment": true, - "reasoning": true, + "inclusionai/ling-2.6-flash": { + "id": "inclusionai/ling-2.6-flash", + "name": "Ling 2.6 Flash", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 262144, + "input": 262144, + "output": 32768 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.08, + "output": 0.24 } }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM-4.5-Air", - "family": "glm-air", + "Alibaba-NLP/Tongyi-DeepResearch-30B-A3B": { + "id": "Alibaba-NLP/Tongyi-DeepResearch-30B-A3B", + "name": "Tongyi DeepResearch 30B A3B", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "yi", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 98304 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 1.1, - "cache_read": 0.03, - "cache_write": 0 + "input": 0.08, + "output": 0.24000000000000002 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, + "ibm-granite/granite-4.1-8b": { + "id": "ibm-granite/granite-4.1-8b", + "name": "Granite 4.1 8B", + "description": "Tool-capable chat model for instruction following and agentic application workflows", + "attachment": false, + "reasoning": false, "tool_call": true, - "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "structured_output": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 131072, + "input": 131072, + "output": 131072 }, "cost": { - "input": 30, - "output": 180 + "input": 0.05, + "output": 0.1, + "cache_read": 0.05 } }, - "gpt-3.5-turbo": { - "id": "gpt-3.5-turbo", - "name": "GPT-3.5-turbo", - "family": "gpt", + "Salesforce/Llama-xLAM-2-70b-fc-r": { + "id": "Salesforce/Llama-xLAM-2-70b-fc-r", + "name": "Llama-xLAM-2 70B fc-r", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "knowledge": "2021-09-01", - "release_date": "2023-03-01", - "last_updated": "2023-11-06", + "release_date": "2025-04-13", + "last_updated": "2025-04-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16385, - "output": 4096 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.5, - "output": 1.5, - "cache_read": 1.25 + "input": 2.5, + "output": 2.5 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o-mini", + "THUDM/GLM-Z1-32B-0414": { + "id": "THUDM/GLM-Z1-32B-0414", + "name": "GLM Z1 32B 0414", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm-z", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.2, + "output": 0.2 } }, - "qwen-vl-max": { - "id": "qwen-vl-max", - "name": "Qwen-VL Max", - "family": "qwen", + "THUDM/GLM-4-32B-0414": { + "id": "THUDM/GLM-4-32B-0414", + "name": "GLM 4 32B 0414", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-04-08", - "last_updated": "2025-08-13", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.8, - "output": 3.2 + "input": 0.2, + "output": 0.2 } }, - "sonar": { - "id": "sonar", - "name": "Sonar", - "family": "sonar", + "THUDM/GLM-4-9B-0414": { + "id": "THUDM/GLM-4-9B-0414", + "name": "GLM 4 9B 0414", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "structured_output": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 32000, + "input": 32000, + "output": 8000 }, "cost": { - "input": 1, - "output": 1 + "input": 0.2, + "output": 0.2 } }, - "qwen3-coder-flash": { - "id": "qwen3-coder-flash", - "name": "Qwen3 Coder Flash", - "family": "qwen", + "THUDM/GLM-Z1-9B-0414": { + "id": "THUDM/GLM-Z1-9B-0414", + "name": "GLM Z1 9B 0414", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm-z", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 65536 + "context": 32000, + "input": 32000, + "output": 8000 }, "cost": { - "input": 0.3, - "output": 1.5 + "input": 0.2, + "output": 0.2 } }, - "grok-4-3": { - "id": "grok-4-3", - "name": "Grok 4.3", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-05-01", - "last_updated": "2026-05-01", + "meta-llama/llama-3.1-8b-instruct": { + "id": "meta-llama/llama-3.1-8b-instruct", + "name": "Llama 3.1 8b Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 30000 + "context": 131072, + "input": 131072, + "output": 16384 }, "cost": { - "input": 1.25, - "output": 2.5, - "cache_read": 0.2, - "context_over_200k": { - "input": 2.5, - "output": 5, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2.5, - "output": 5, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.0544, + "output": 0.0544 } }, - "glm-4.5v": { - "id": "glm-4.5v", - "name": "GLM-4.5V", - "family": "glm", + "meta-llama/llama-4-maverick": { + "id": "meta-llama/llama-4-maverick", + "name": "Llama 4 Maverick", + "description": "Open multimodal Llama model for strong reasoning and fast responses", + "family": "llama", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", + "structured_output": true, + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 64000, - "output": 16384 + "context": 1048576, + "input": 1048576, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 1.8 + "input": 0.18000000000000002, + "output": 0.8 } }, - "deepseek-v4-flash": { - "id": "deepseek-v4-flash", - "name": "DeepSeek V4 Flash", - "family": "deepseek-flash", + "meta-llama/llama-3.3-70b-instruct": { + "id": "meta-llama/llama-3.3-70b-instruct", + "name": "Llama 3.3 70b Instruct", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 131072, + "input": 131072, + "output": 16384 }, "cost": { - "input": 0.14, - "output": 0.28, - "cache_read": 0.028 + "input": 0.05, + "output": 0.23 } }, - "grok-4": { - "id": "grok-4", - "name": "Grok 4", - "family": "grok", - "attachment": false, - "reasoning": true, + "meta-llama/llama-4-scout": { + "id": "meta-llama/llama-4-scout", + "name": "Llama 4 Scout", + "description": "Open multimodal Llama model for long-context analysis and efficient agents", + "family": "llama", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "structured_output": true, + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 328000, + "input": 328000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "reasoning": 15, - "cache_read": 0.75 + "input": 0.085, + "output": 0.46 } }, - "qwen3-next-80b-a3b-instruct": { - "id": "qwen3-next-80b-a3b-instruct", - "name": "Qwen3-Next 80B-A3B Instruct", - "family": "qwen", - "attachment": false, + "meta-llama/llama-3.2-3b-instruct": { + "id": "meta-llama/llama-3.2-3b-instruct", + "name": "Llama 3.2 3b Instruct", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09", - "last_updated": "2025-09", + "tool_call": false, + "structured_output": false, + "release_date": "2024-09-25", + "last_updated": "2024-09-25", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, - "output": 32768 + "input": 131072, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 2 + "input": 0.0306, + "output": 0.0493 } }, - "gpt-4": { - "id": "gpt-4", - "name": "GPT-4", - "family": "gpt", - "attachment": true, + "featherless-ai/Qwerky-72B": { + "id": "featherless-ai/Qwerky-72B", + "name": "Qwerky 72B", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "qwerky", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2025-03-20", + "last_updated": "2025-03-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, + "context": 32000, + "input": 32000, "output": 8192 }, "cost": { - "input": 30, - "output": 60 + "input": 0.5, + "output": 0.5 } }, - "glm-4.6": { - "id": "glm-4.6", - "name": "GLM-4.6", - "family": "glm", + "moonshotai/kimi-k2-instruct-0711": { + "id": "moonshotai/kimi-k2-instruct-0711", + "name": "Kimi K2 0711", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "structured_output": true, + "release_date": "2025-07-11", + "last_updated": "2025-07-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "input": 0.1, + "output": 2 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", - "attachment": true, - "reasoning": true, + "moonshotai/Kimi-K2-Instruct-0905": { + "id": "moonshotai/Kimi-K2-Instruct-0905", + "name": "Kimi K2 0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 256000, + "input": 256000, "output": 262144 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.4, + "output": 2 } }, - "glm-4.6v": { - "id": "glm-4.6v", - "name": "GLM-4.6V", - "family": "glm", - "attachment": true, + "moonshotai/kimi-k2-thinking-original": { + "id": "moonshotai/kimi-k2-thinking-original", + "name": "Kimi K2 Thinking Original", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 0.9 + "input": 0.6, + "output": 2.5 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "moonshotai/kimi-k2.5:thinking": { + "id": "moonshotai/kimi-k2.5:thinking", + "name": "Kimi K2.5 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": false, + "release_date": "2026-01-26", + "last_updated": "2026-01-26", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 256000, + "input": 256000, + "output": 65536 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.3, + "output": 1.9 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, - "reasoning": true, + "moonshotai/kimi-k2-instruct": { + "id": "moonshotai/kimi-k2-instruct", + "name": "Kimi K2 Instruct", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 + "input": 0.1, + "output": 2 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, - "reasoning": true, + "moonshotai/kimi-k2-thinking": { + "id": "moonshotai/kimi-k2-thinking", + "name": "Kimi K2 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "structured_output": true, + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.3, + "output": 1.2 } }, - "glm-4.5-flash": { - "id": "glm-4.5-flash", - "name": "GLM-4.5-Flash", - "family": "glm-flash", - "attachment": false, + "moonshotai/kimi-k2.6:thinking": { + "id": "moonshotai/kimi-k2.6:thinking", + "name": "Kimi K2.6 Thinking", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "structured_output": false, + "release_date": "2026-04-16", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.53, + "output": 2.73 } }, - "qwen3-vl-plus": { - "id": "qwen3-vl-plus", - "name": "Qwen3-VL Plus", - "family": "qwen", - "attachment": false, - "reasoning": true, + "moonshotai/kimi-k2.5": { + "id": "moonshotai/kimi-k2.5", + "name": "Kimi K2.5", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-23", - "last_updated": "2025-09-23", + "structured_output": false, + "release_date": "2026-01-26", + "last_updated": "2026-01-26", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 32768 + "context": 256000, + "input": 256000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 1.6, - "reasoning": 4.8 + "input": 0.3, + "output": 1.9 } }, - "grok-4-1-fast": { - "id": "grok-4-1-fast", - "name": "Grok 4.1 Fast", - "family": "grok", + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "family": "kimi-k2", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "structured_output": false, + "release_date": "2026-04-16", + "last_updated": "2026-04-21", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 2000000, - "output": 30000 + "context": 256000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.53, + "output": 2.73 } }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "moonshotai/kimi-latest": { + "id": "moonshotai/kimi-latest", + "name": "Kimi Latest", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": false, + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 256000, + "input": 256000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.5, + "output": 2.6, + "cache_read": 0.125 } }, - "qwen3-coder-480b-a35b-instruct": { - "id": "qwen3-coder-480b-a35b-instruct", - "name": "Qwen3-Coder 480B-A35B Instruct", - "family": "qwen", + "moonshotai/kimi-k2-thinking-turbo-original": { + "id": "moonshotai/kimi-k2-thinking-turbo-original", + "name": "Kimi K2 Thinking Turbo Original", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-06", + "last_updated": "2025-11-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 16384 }, "cost": { - "input": 1.5, - "output": 7.5 + "input": 1.15, + "output": 8 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "baidu/ernie-4.5-vl-28b-a3b": { + "id": "baidu/ernie-4.5-vl-28b-a3b", + "name": "ERNIE 4.5 VL 28B", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "ernie", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-06-30", + "last_updated": "2025-06-30", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 128000 + "context": 32768, + "input": 32768, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.13999999999999999, + "output": 0.5599999999999999 } }, - "deepseek-v4-pro": { - "id": "deepseek-v4-pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek-thinking", - "attachment": false, + "perceptron/perceptron-mk1": { + "id": "perceptron/perceptron-mk1", + "name": "Perceptron Mk1", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [], + "tool_call": false, "structured_output": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 384000 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 1.74, - "output": 3.48, - "cache_read": 0.145 + "input": 0.15, + "output": 1.5 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "GPT-4.1 nano", - "family": "gpt-nano", - "attachment": true, + "failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5": { + "id": "failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5", + "name": "Llama 3 70B abliterated", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.03 + "input": 0.7, + "output": 0.7 } }, - "claude-3-7-sonnet-20250219": { - "id": "claude-3-7-sonnet-20250219", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", - "attachment": true, + "nanogpt/coding-router:max": { + "id": "nanogpt/coding-router:max", + "name": "Coding Router Max", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "structured_output": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "qwen3-coder-30b-a3b-instruct": { - "id": "qwen3-coder-30b-a3b-instruct", - "name": "Qwen3-Coder 30B-A3B Instruct", - "family": "qwen", + "nanogpt/coding-router:high": { + "id": "nanogpt/coding-router:high", + "name": "Coding Router High", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "structured_output": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.45, - "output": 2.25 + "input": 1.1, + "output": 2.2, + "cache_read": 0.11 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "nanogpt/coding-router:low": { + "id": "nanogpt/coding-router:low", + "name": "Coding Router Low", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "structured_output": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "mimo-v2-pro": { - "id": "mimo-v2-pro", - "name": "MiMo-V2-Pro", - "family": "mimo", + "nanogpt/coding-router:medium": { + "id": "nanogpt/coding-router:medium", + "name": "Coding Router Medium", + "description": "Automatic model router for matching prompts to suitable backends and budgets", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ] + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "o3": { - "id": "o3", - "name": "o3", - "family": "o", - "attachment": true, + "nanogpt/coding-router": { + "id": "nanogpt/coding-router", + "name": "Coding Router", + "description": "Automatic model router for matching prompts to suitable backends and budgets", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 1.1, + "output": 2.2, + "cache_read": 0.11 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "GalrionSoftworks/MN-LooseCannon-12B-v1": { + "id": "GalrionSoftworks/MN-LooseCannon-12B-v1", + "name": "MN-LooseCannon-12B-v1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 16384, + "input": 16384, + "output": 8192 + }, + "cost": { + "input": 0.49299999999999994, + "output": 0.49299999999999994 + } + }, + "TheDrummer/Cydonia-24B-v4.3": { + "id": "TheDrummer/Cydonia-24B-v4.3", + "name": "The Drummer Cydonia 24B v4.3", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-25", + "last_updated": "2025-12-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 272000 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 15, - "output": 120 + "input": 0.1003, + "output": 0.1207 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, + "TheDrummer/Cydonia-24B-v4.1": { + "id": "TheDrummer/Cydonia-24B-v4.1", + "name": "The Drummer Cydonia 24B v4.1", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-19", + "last_updated": "2025-08-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.1003, + "output": 0.1207 } }, - "minimax-m2.5-highspeed": { - "id": "minimax-m2.5-highspeed", - "name": "MiniMax-M2.5-highspeed", - "family": "minimax", + "TheDrummer/Cydonia-24B-v2": { + "id": "TheDrummer/Cydonia-24B-v2", + "name": "The Drummer Cydonia 24B v2", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-17", + "last_updated": "2025-02-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.1003, + "output": 0.1207 } }, - "qwen-turbo": { - "id": "qwen-turbo", - "name": "Qwen Turbo", - "family": "qwen", + "TheDrummer/Anubis-70B-v1.1": { + "id": "TheDrummer/Anubis-70B-v1.1", + "name": "Anubis 70B v1.1", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-11-01", - "last_updated": "2025-04-28", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 131072, + "input": 131072, "output": 16384 }, "cost": { - "input": 0.05, - "output": 0.2, - "reasoning": 0.5 + "input": 0.31, + "output": 0.31 } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5 (latest)", - "family": "claude-sonnet", + "TheDrummer/UnslopNemo-12B-v4.1": { + "id": "TheDrummer/UnslopNemo-12B-v4.1", + "name": "UnslopNemo 12b v4", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.493, + "output": 0.493 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", + "TheDrummer/skyfall-36b-v2": { + "id": "TheDrummer/skyfall-36b-v2", + "name": "TheDrummer Skyfall 36B V2", + "description": "Multimodal model for analyzing text, images, documents, and rich media", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 64000, + "input": 64000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.01, - "input_audio": 0.3 + "input": 0.493, + "output": 0.493 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "TheDrummer/Rocinante-12B-v1.1": { + "id": "TheDrummer/Rocinante-12B-v1.1", + "name": "Rocinante 12b", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.408, + "output": 0.595 } }, - "glm-4.7-flash": { - "id": "glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + "TheDrummer/Magidonia-24B-v4.3": { + "id": "TheDrummer/Magidonia-24B-v4.3", + "name": "The Drummer Magidonia 24B v4.3", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-25", + "last_updated": "2025-12-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 131072 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.1003, + "output": 0.1207 } }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "MiMo-V2-Flash", - "family": "mimo", + "TheDrummer/Cydonia-24B-v4": { + "id": "TheDrummer/Cydonia-24B-v4", + "name": "The Drummer Cydonia 24B v4", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-22", + "last_updated": "2025-07-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 0.2006, + "output": 0.2414 } }, - "qwen3.6-max-preview": { - "id": "qwen3.6-max-preview", - "name": "Qwen3.6 Max Preview", - "family": "qwen", + "TheDrummer/Skyfall-31B-v4.2": { + "id": "TheDrummer/Skyfall-31B-v4.2", + "name": "TheDrummer Skyfall 31B v4.2", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-26", + "last_updated": "2026-03-26", + "modalities": { + "input": ["text", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "input": 131072, + "output": 16384 + }, + "cost": { + "input": 0.55, + "output": 0.8 + } + }, + "TheDrummer/Anubis-70B-v1": { + "id": "TheDrummer/Anubis-70B-v1", + "name": "Anubis 70B v1", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 65536 + "context": 65536, + "input": 65536, + "output": 16384 }, "cost": { - "input": 1.3, - "output": 7.8, - "cache_read": 0.13, - "cache_write": 1.625 + "input": 0.31, + "output": 0.31 } }, - "gpt-5-chat-latest": { - "id": "gpt-5-chat-latest", - "name": "GPT-5 Chat (latest)", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, + "huihui-ai/Qwen2.5-32B-Instruct-abliterated": { + "id": "huihui-ai/Qwen2.5-32B-Instruct-abliterated", + "name": "Qwen 2.5 32B Abliterated", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": false, + "release_date": "2025-01-06", + "last_updated": "2025-01-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10 + "input": 0.7, + "output": 0.7 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "huihui-ai/Llama-3.3-70B-Instruct-abliterated": { + "id": "huihui-ai/Llama-3.3-70B-Instruct-abliterated", + "name": "Llama 3.3 70B Instruct abliterated", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.7, + "output": 0.7 } }, - "qwen2-5-vl-72b-instruct": { - "id": "qwen2-5-vl-72b-instruct", - "name": "Qwen2.5-VL 72B Instruct", + "huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated": { + "id": "huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated", + "name": "DeepSeek R1 Qwen Abliterated", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-09", - "last_updated": "2024-09", + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 16384, + "input": 16384, "output": 8192 }, "cost": { - "input": 2.8, - "output": 8.4 + "input": 1.4, + "output": 1.4 } }, - "gpt-5.5-pro": { - "id": "gpt-5.5-pro", - "name": "GPT-5.5 Pro", - "family": "gpt-pro", - "attachment": true, + "huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated": { + "id": "huihui-ai/DeepSeek-R1-Distill-Llama-70B-abliterated", + "name": "DeepSeek R1 Llama 70B Abliterated", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-23", - "last_updated": "2026-04-23", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 30, - "output": 180, - "context_over_200k": { - "input": 60, - "output": 270 - }, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.7, + "output": 0.7 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B": { + "id": "Envoid/Llama-3.05-NT-Storybreaker-Ministral-70B", + "name": "Llama 3.05 Storybreaker Ministral 70b", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "devstral-small-2507": { - "id": "devstral-small-2507", - "name": "Devstral Small", - "family": "devstral", + "Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B": { + "id": "Envoid/Llama-3.05-Nemotron-Tenyxchat-Storybreaker-70B", + "name": "Nemotron Tenyxchat Storybreaker 70b", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-07-10", - "last_updated": "2025-07-10", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "stepfun-ai/step-3.5-flash-2603": { + "id": "stepfun-ai/step-3.5-flash-2603", + "name": "Step 3.5 Flash 2603", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", "attachment": false, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-14", + "last_updated": "2026-04-14", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.1, + "output": 0.3 } }, - "gemini-2.0-flash-lite": { - "id": "gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "stepfun-ai/step-3.5-flash": { + "id": "stepfun-ai/step-3.5-flash", + "name": "Step 3.5 Flash", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "family": "step", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-02", + "last_updated": "2026-02-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.2, + "output": 0.5 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", + "mistral/mistral-medium-3.5": { + "id": "mistral/mistral-medium-3.5", + "name": "Mistral Medium 3.5", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, + "context": 256000, + "input": 256000, "output": 32768 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 1.5, + "output": 7.5 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "mistral/mistral-medium-3.5:thinking": { + "id": "mistral/mistral-medium-3.5:thinking", + "name": "Mistral Medium 3.5 Thinking", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 1.5, + "output": 7.5 } }, - "grok-3": { - "id": "grok-3", - "name": "Grok 3", - "family": "grok", + "Tongyi-Zhiwen/QwenLong-L1-32B": { + "id": "Tongyi-Zhiwen/QwenLong-L1-32B", + "name": "QwenLong L1 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-25", + "last_updated": "2025-01-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 40960 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.13999999999999999, + "output": 0.6 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, + "google/gemini-flash-1.5": { + "id": "google/gemini-flash-1.5", + "name": "Gemini 1.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-14", + "last_updated": "2024-05-14", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 2000000, - "output": 30000 + "input": 2000000, + "output": 8192 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.0748, + "output": 0.306 } }, - "sonar-reasoning-pro": { - "id": "sonar-reasoning-pro", - "name": "Sonar Reasoning Pro", - "family": "sonar-reasoning", + "google/gemini-3.1-pro-preview-high": { + "id": "google/gemini-3.1-pro-preview-high", + "name": "Gemini 3.1 Pro (Preview High)", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": true, - "knowledge": "2025-09-01", - "release_date": "2024-01-01", - "last_updated": "2025-09-01", + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-21", + "last_updated": "2026-02-21", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { "input": 2, - "output": 8 + "output": 12, + "cache_read": 0.2 } - } - } - }, - "google-vertex": { - "id": "google-vertex", - "env": ["GOOGLE_VERTEX_PROJECT", "GOOGLE_VERTEX_LOCATION", "GOOGLE_APPLICATION_CREDENTIALS"], - "npm": "@ai-sdk/google-vertex", - "name": "Vertex", - "doc": "https://cloud.google.com/vertex-ai/generative-ai/docs/models", - "models": { - "gemini-2.0-flash": { - "id": "gemini-2.0-flash", - "name": "Gemini 2.0 Flash", - "family": "gemini-flash", + }, + "google/gemini-3.1-flash-lite": { + "id": "google/gemini-3.1-flash-lite", + "name": "Gemini 3.1 Flash Lite", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": true, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 8192 + "input": 1048576, + "output": 65536 }, "cost": { - "input": 0.15, - "output": 0.6, + "input": 0.25, + "output": 1.5, "cache_read": 0.025 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", + "google/gemini-3-flash-preview-thinking": { + "id": "google/gemini-3-flash-preview-thinking", + "name": "Gemini 3 Flash Thinking", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1048756, + "input": 1048756, "output": 65536 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.5, + "output": 3 } }, - "gemini-flash-latest": { - "id": "gemini-flash-latest", - "name": "Gemini Flash Latest", - "family": "gemini-flash", + "google/gemini-3.5-flash": { + "id": "google/gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "structured_output": true, + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "audio"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, + "input": 1048576, "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "cache_write": 0.383 + "input": 1.5, + "output": 9, + "cache_read": 0.15 } }, - "gemini-2.5-flash-lite-preview-06-17": { - "id": "gemini-2.5-flash-lite-preview-06-17", - "name": "Gemini 2.5 Flash Lite Preview 06-17", - "family": "gemini-flash-lite", + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 65536 + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "output": 0.35 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", + "google/gemini-pro-latest": { + "id": "google/gemini-pro-latest", + "name": "Gemini Pro Latest", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "structured_output": true, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1048756, + "input": 1048756, "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "cache_write": 0.383 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "gemini-2.5-flash-preview-09-2025": { - "id": "gemini-2.5-flash-preview-09-2025", - "name": "Gemini 2.5 Flash Preview 09-25", - "family": "gemini-flash", + "google/gemini-3.1-pro-preview-customtools": { + "id": "google/gemini-3.1-pro-preview-customtools", + "name": "Gemini 3.1 Pro (Preview Custom Tools)", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "structured_output": true, + "release_date": "2026-02-27", + "last_updated": "2026-02-27", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 1048756, + "input": 1048756, "output": 65536 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "cache_write": 0.383 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "zai-org/glm-5-maas": { - "id": "zai-org/glm-5-maas", - "name": "GLM-5", - "family": "glm", - "attachment": false, + "google/gemini-flash-lite-latest": { + "id": "google/gemini-flash-lite-latest", + "name": "Gemini Flash Lite Latest", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "structured_output": true, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202752, - "output": 131072 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 1048576, + "input": 1048576, + "output": 65536 }, "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.1 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025 } }, - "zai-org/glm-4.7-maas": { - "id": "zai-org/glm-4.7-maas", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, + "google/gemma-4-26b-a4b-it:thinking": { + "id": "google/gemma-4-26b-a4b-it:thinking", + "name": "Gemma 4 26B A4B Thinking", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "reasoning_options": [], + "tool_call": false, "structured_output": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-06", - "last_updated": "2026-01-06", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 128000 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 2.2 + "input": 0.13, + "output": 0.4 } }, - "deepseek-ai/deepseek-v3.2-maas": { - "id": "deepseek-ai/deepseek-v3.2-maas", - "name": "DeepSeek V3.2", - "family": "deepseek", - "attachment": false, + "google/gemini-3.1-pro-preview": { + "id": "google/gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro (Preview)", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-12-17", - "last_updated": "2026-04-04", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, + "context": 1048756, + "input": 1048756, "output": 65536 }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" - }, "cost": { - "input": 0.56, - "output": 1.68, - "cache_read": 0.056 + "input": 2, + "output": 12, + "cache_read": 0.2 } }, - "deepseek-ai/deepseek-v3.1-maas": { - "id": "deepseek-ai/deepseek-v3.1-maas", - "name": "DeepSeek V3.1", - "family": "deepseek", - "attachment": false, + "google/gemma-4-26b-a4b-it": { + "id": "google/gemma-4-26b-a4b-it", + "name": "Gemma 4 26B A4B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, "structured_output": true, - "temperature": true, - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 32768 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 1.7 + "input": 0.13, + "output": 0.4 } }, - "openai/gpt-oss-120b-maas": { - "id": "openai/gpt-oss-120b-maas", - "name": "GPT OSS 120B", - "family": "gpt-oss", - "attachment": false, + "google/gemini-3.1-pro-preview-low": { + "id": "google/gemini-3.1-pro-preview-low", + "name": "Gemini 3.1 Pro (Preview Low)", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "release_date": "2026-02-21", + "last_updated": "2026-02-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1048756, + "input": 1048756, + "output": 65536 + }, + "cost": { + "input": 2, + "output": 12, + "cache_read": 0.2 + } + }, + "google/gemma-4-31b-it:thinking": { + "id": "google/gemma-4-31b-it:thinking", + "name": "Gemma 4 31B Thinking", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 0.09, - "output": 0.36 + "input": 0.1, + "output": 0.35 } }, - "openai/gpt-oss-20b-maas": { - "id": "openai/gpt-oss-20b-maas", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, + "google/gemini-3.5-flash-thinking": { + "id": "google/gemini-3.5-flash-thinking", + "name": "Gemini 3.5 Flash Thinking", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1048576, + "input": 1048576, + "output": 65536 }, "cost": { - "input": 0.07, - "output": 0.25 + "input": 1.5, + "output": 9, + "cache_read": 0.15 } }, - "meta/llama-3.3-70b-instruct-maas": { - "id": "meta/llama-3.3-70b-instruct-maas", - "name": "Llama 3.3 70B Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "google/gemini-3-flash-preview": { + "id": "google/gemini-3-flash-preview", + "name": "Gemini 3 Flash (Preview)", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.72, - "output": 0.72 + "input": 0.5, + "output": 3 } }, - "meta/llama-4-maverick-17b-128e-instruct-maas": { - "id": "meta/llama-4-maverick-17b-128e-instruct-maas", - "name": "Llama 4 Maverick 17B 128E Instruct", - "family": "llama", + "google/gemini-flash-latest": { + "id": "google/gemini-flash-latest", + "name": "Gemini Flash Latest", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-29", - "last_updated": "2025-04-29", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 524288, - "output": 8192 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 1048756, + "input": 1048756, + "output": 65536 }, "cost": { - "input": 0.35, - "output": 1.15 + "input": 1.5, + "output": 9, + "cache_read": 0.15 } }, - "qwen/qwen3-235b-a22b-instruct-2507-maas": { - "id": "qwen/qwen3-235b-a22b-instruct-2507-maas", - "name": "Qwen3 235B A22B Instruct", - "family": "qwen", + "liquid/lfm-2-24b-a2b": { + "id": "liquid/lfm-2-24b-a2b", + "name": "LFM2 24B A2B", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-13", - "last_updated": "2025-08-13", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-20", + "last_updated": "2025-12-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 16384 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.22, - "output": 0.88 + "input": 0.03, + "output": 0.12 } }, - "moonshotai/kimi-k2-thinking-maas": { - "id": "moonshotai/kimi-k2-thinking-maas", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", - "attachment": false, + "x-ai/grok-4.20": { + "id": "x-ai/grok-4.20", + "name": "Grok 4.20", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-03-31", + "last_updated": "2026-03-31", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi" + "context": 2000000, + "input": 2000000, + "output": 131072 }, "cost": { - "input": 0.6, - "output": 2.5 + "input": 2, + "output": 6 } }, - "gemini-flash-lite-latest": { - "id": "gemini-flash-lite-latest", - "name": "Gemini Flash-Lite Latest", - "family": "gemini-flash-lite", + "x-ai/grok-4.3": { + "id": "x-ai/grok-4.3", + "name": "Grok 4.3", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "structured_output": true, + "release_date": "2026-04-30", + "last_updated": "2026-04-30", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 1000000, + "input": 1000000, + "output": 1000000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "gemini-2.5-pro-preview-05-06": { - "id": "gemini-2.5-pro-preview-05-06", - "name": "Gemini 2.5 Pro Preview 05-06", - "family": "gemini-pro", + "x-ai/grok-4.20-multi-agent": { + "id": "x-ai/grok-4.20-multi-agent", + "name": "Grok 4.20 Multi-Agent", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-05-06", - "last_updated": "2025-05-06", + "structured_output": true, + "release_date": "2026-03-31", + "last_updated": "2026-03-31", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 2000000, + "input": 2000000, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 2, + "output": 6 } }, - "claude-haiku-4-5@20251001": { - "id": "claude-haiku-4-5@20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "x-ai/grok-latest": { + "id": "x-ai/grok-latest", + "name": "Grok Latest", + "description": "Grok model for agentic tool use, reasoning, coding, and live assistance", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "structured_output": true, + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 1000000, + "input": 1000000, + "output": 1000000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 1.25, + "output": 2.5, + "cache_read": 0.2 } }, - "gemini-3.1-pro-preview-customtools": { - "id": "gemini-3.1-pro-preview-customtools", - "name": "Gemini 3.1 Pro Preview Custom Tools", - "family": "gemini-pro", + "x-ai/grok-build-0.1": { + "id": "x-ai/grok-build-0.1", + "name": "Grok Build 0.1", + "description": "Grok coding model for agentic engineering, edits, and codebase workflows", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "release_date": "2026-05-20", + "last_updated": "2026-05-20", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 1, + "output": 2, + "cache_read": 0.2 } }, - "claude-sonnet-4-6@default": { - "id": "claude-sonnet-4-6@default", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0": { + "id": "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.0", + "name": "EVA Llama 3.33 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75, - "context_over_200k": { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5 - }, - "tiers": [ - { - "input": 6, - "output": 22.5, - "cache_read": 0.6, - "cache_write": 7.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 2.006, + "output": 2.006 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview 09-25", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", + "EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2": { + "id": "EVA-UNIT-01/EVA-Qwen2.5-72B-v0.2", + "name": "EVA-Qwen2.5-72B-v0.2", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, "release_date": "2025-09-25", "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.7989999999999999, + "output": 0.7989999999999999 } }, - "claude-3-5-haiku@20241022": { - "id": "claude-3-5-haiku@20241022", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", - "attachment": true, + "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1": { + "id": "EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1", + "name": "EVA-LLaMA-3.33-70B-v0.1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 8192 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 2.006, + "output": 2.006 } }, - "gemini-3.1-flash-lite-preview": { - "id": "gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2": { + "id": "EVA-UNIT-01/EVA-Qwen2.5-32B-v0.2", + "name": "EVA-Qwen2.5-32B-v0.2", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 0.7989999999999999, + "output": 0.7989999999999999 } }, - "claude-3-5-sonnet@20241022": { - "id": "claude-3-5-sonnet@20241022", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", + "microsoft/wizardlm-2-8x22b": { + "id": "microsoft/wizardlm-2-8x22b", + "name": "WizardLM-2 8x22B", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, + "context": 65536, + "input": 65536, "output": 8192 }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" - }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "Doctor-Shotgun/MS3.2-24B-Magnum-Diamond": { + "id": "Doctor-Shotgun/MS3.2-24B-Magnum-Diamond", + "name": "MS3.2 24B Magnum Diamond", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - }, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "claude-opus-4-1@20250805": { - "id": "claude-opus-4-1@20250805", - "name": "Claude Opus 4.1", - "family": "claude-opus", - "attachment": true, + "poolside/laguna-xs.2": { + "id": "poolside/laguna-xs.2", + "name": "Laguna XS.2", + "description": "Agentic coding model from Poolside in the XS size class for local deployment", + "family": "laguna", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-04-28", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 262144, + "output": 32768 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.2, + "output": 0.4 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", - "attachment": true, + "poolside/laguna-m.1": { + "id": "poolside/laguna-m.1", + "name": "Laguna M.1", + "description": "Poolside's flagship agentic coding model for long-horizon work", + "family": "laguna", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, + "structured_output": false, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2026-04-28", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 3, - "cache_read": 0.05, - "input_audio": 1 + "input": 0.2, + "output": 0.4 } }, - "gemini-2.5-flash-preview-05-20": { - "id": "gemini-2.5-flash-preview-05-20", - "name": "Gemini 2.5 Flash Preview 05-20", - "family": "gemini-flash", + "z-ai/glm-4.5v:thinking": { + "id": "z-ai/glm-4.5v:thinking", + "name": "GLM 4.5V Thinking", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-22", + "last_updated": "2025-11-22", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 64000, + "input": 64000, + "output": 96000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.0375 + "input": 0.6, + "output": 1.7999999999999998 } }, - "gemini-embedding-001": { - "id": "gemini-embedding-001", - "name": "Gemini Embedding 001", - "family": "gemini", + "z-ai/glm-4.6:thinking": { + "id": "z-ai/glm-4.6:thinking", + "name": "GLM 4.6 Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "knowledge": "2025-05", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2048, - "output": 3072 + "context": 200000, + "input": 200000, + "output": 65535 }, "cost": { - "input": 0.15, - "output": 0 + "input": 0.4, + "output": 1.5 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", + "z-ai/glm-4.5v": { + "id": "z-ai/glm-4.5v", + "name": "GLM 4.5V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glmv", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-22", + "last_updated": "2025-11-22", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 64000, + "input": 64000, + "output": 96000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - }, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 0.6, + "output": 1.7999999999999998 } }, - "gemini-2.5-pro-preview-06-05": { - "id": "gemini-2.5-pro-preview-06-05", - "name": "Gemini 2.5 Pro Preview 06-05", - "family": "gemini-pro", - "attachment": true, + "z-ai/glm-4.6": { + "id": "z-ai/glm-4.6", + "name": "GLM 4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "structured_output": true, + "release_date": "2025-09-30", + "last_updated": "2025-09-30", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 65535 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 0.4, + "output": 1.5 } }, - "claude-sonnet-4@20250514": { - "id": "claude-sonnet-4@20250514", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", + "z-ai/glm-5v-turbo:thinking": { + "id": "z-ai/glm-5v-turbo:thinking", + "name": "GLM 5V Turbo Thinking", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": true, + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 202800, + "input": 202800, + "output": 131100 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "gemini-3.1-flash-lite": { - "id": "gemini-3.1-flash-lite", - "name": "Gemini 3.1 Flash Lite", - "family": "gemini-flash-lite", + "z-ai/glm-5v-turbo": { + "id": "z-ai/glm-5v-turbo", + "name": "GLM 5V Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-05-07", - "last_updated": "2026-05-07", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 202800, + "input": 202800, + "output": 131100 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "claude-3-7-sonnet@20250219": { - "id": "claude-3-7-sonnet@20250219", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, + "z-ai/glm-5-turbo": { + "id": "z-ai/glm-5-turbo", + "name": "GLM 5 Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "structured_output": true, + "release_date": "2026-03-15", + "last_updated": "2026-03-15", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 202800, + "input": 202800, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.2, + "output": 4, + "cache_read": 0.24 } }, - "claude-opus-4@20250514": { - "id": "claude-opus-4@20250514", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, + "openai/o3-mini-low": { + "id": "openai/o3-mini-low", + "name": "OpenAI o3-mini (Low)", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "structured_output": true, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 32000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "input": 200000, + "output": 100000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 9.996, + "output": 19.992 } }, - "claude-opus-4-5@20251101": { - "id": "claude-opus-4-5@20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "openai/gpt-oss-safeguard-20b": { + "id": "openai/gpt-oss-safeguard-20b", + "name": "GPT OSS Safeguard 20B", + "description": "Safety model for policy screening, moderation, and risk-aware routing workflows", + "family": "gpt-oss", + "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-10-29", + "last_updated": "2025-10-29", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.075, + "output": 0.3 } }, - "gemini-2.5-flash-preview-04-17": { - "id": "gemini-2.5-flash-preview-04-17", - "name": "Gemini 2.5 Flash Preview 04-17", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-04-17", - "last_updated": "2025-04-17", + "openai/o3": { + "id": "openai/o3", + "name": "OpenAI o3", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.0375 + "input": 2, + "output": 8 } }, - "claude-sonnet-4-5@20250929": { - "id": "claude-sonnet-4-5@20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "openai/o4-mini-high": { + "id": "openai/o4-mini-high", + "name": "OpenAI o4-mini high", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "structured_output": true, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" + "input": 200000, + "output": 100000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.1, + "output": 4.4 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, + "openai/o3-pro-2025-06-10": { + "id": "openai/o3-pro-2025-06-10", + "name": "OpenAI o3-pro (2025-06-10)", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "structured_output": true, + "release_date": "2025-06-10", + "last_updated": "2025-06-10", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.01, - "input_audio": 0.3 + "input": 9.996, + "output": 19.992 } }, - "claude-opus-4-6@default": { - "id": "claude-opus-4-6@default", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "openai/gpt-5.2-pro": { + "id": "openai/gpt-5.2-pro", + "name": "GPT 5.2 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "structured_output": true, + "release_date": "2026-01-01", + "last_updated": "2026-01-01", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, + "input": 400000, "output": 128000 }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" - }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - }, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 21, + "output": 168 } }, - "gemini-2.0-flash-lite": { - "id": "gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, + "openai/gpt-4o-mini-search-preview": { + "id": "openai/gpt-4o-mini-search-preview", + "name": "GPT-4o mini Search Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.088, + "output": 0.35 } }, - "claude-opus-4-7@default": { - "id": "claude-opus-4-7@default", - "name": "Claude Opus 4.7", - "family": "claude-opus", + "openai/gpt-5": { + "id": "openai/gpt-5", + "name": "GPT 5", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "structured_output": true, + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 400000, + "input": 400000, "output": 128000 }, - "provider": { - "npm": "@ai-sdk/google-vertex/anthropic" - }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - }, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ] + "input": 1.25, + "output": 10 } - } - } - }, - "cloudflare-workers-ai": { - "id": "cloudflare-workers-ai", - "env": ["CLOUDFLARE_ACCOUNT_ID", "CLOUDFLARE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.cloudflare.com/client/v4/accounts/${CLOUDFLARE_ACCOUNT_ID}/ai/v1", - "name": "Cloudflare Workers AI", - "doc": "https://developers.cloudflare.com/workers-ai/models/", - "models": { - "@cf/zai-org/glm-4.7-flash": { - "id": "@cf/zai-org/glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + }, + "openai/gpt-3.5-turbo": { + "id": "openai/gpt-3.5-turbo", + "name": "GPT-3.5 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2022-11-30", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 16385, + "input": 16385, + "output": 4096 }, "cost": { - "input": 0.06, - "output": 0.4 + "input": 0.5, + "output": 1.5 } }, - "@cf/nvidia/nemotron-3-120b-a12b": { - "id": "@cf/nvidia/nemotron-3-120b-a12b", - "name": "Nemotron 3 Super 120B", - "family": "nemotron", - "attachment": false, + "openai/gpt-5-pro": { + "id": "openai/gpt-5-pro", + "name": "GPT 5 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "family": "gpt-pro", + "attachment": true, "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-03-11", - "last_updated": "2026-03-11", + "reasoning_options": [ + { + "type": "effort", + "values": ["high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 15, + "output": 120 } }, - "@cf/openai/gpt-oss-20b": { - "id": "@cf/openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "openai/gpt-4o": { + "id": "openai/gpt-4o", + "name": "GPT-4o", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 16384 }, "cost": { - "input": 0.2, - "output": 0.3 + "input": 2.499, + "output": 9.996 } }, - "@cf/openai/gpt-oss-120b": { - "id": "@cf/openai/gpt-oss-120b", - "name": "GPT OSS 120B", + "openai/o4-mini": { + "id": "openai/o4-mini", + "name": "OpenAI o4-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.35, - "output": 0.75 + "input": 1.1, + "output": 4.4 } }, - "@cf/meta/llama-4-scout-17b-16e-instruct": { - "id": "@cf/meta/llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", + "openai/gpt-5.4-nano": { + "id": "openai/gpt-5.4-nano", + "name": "GPT 5.4 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "structured_output": true, + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0.27, - "output": 0.85 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "@cf/google/gemma-4-26b-a4b-it": { - "id": "@cf/google/gemma-4-26b-a4b-it", - "name": "Gemma 4 26B A4B IT", - "family": "gemma", + "openai/gpt-5.1-codex": { + "id": "openai/gpt-5.1-codex", + "name": "GPT 5.1 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-12-15", - "last_updated": "2025-12-15", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 16384 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 1.25, + "output": 10 } }, - "@cf/moonshotai/kimi-k2.5": { - "id": "@cf/moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "openai/gpt-5.1-codex-max": { + "id": "openai/gpt-5.1-codex-max", + "name": "GPT 5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 2.5, + "output": 20 } }, - "@cf/moonshotai/kimi-k2.6": { - "id": "@cf/moonshotai/kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", + "openai/gpt-4o-2024-08-06": { + "id": "openai/gpt-4o-2024-08-06", + "name": "GPT-4o (2024-08-06)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-08-06", + "last_updated": "2024-08-06", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 2.499, + "output": 9.996 } - } - } - }, - "groq": { - "id": "groq", - "env": ["GROQ_API_KEY"], - "npm": "@ai-sdk/groq", - "name": "Groq", - "doc": "https://console.groq.com/docs/models", - "models": { - "gemma2-9b-it": { - "id": "gemma2-9b-it", - "name": "Gemma 2 9B", - "family": "gemma", + }, + "openai/o1-preview": { + "id": "openai/o1-preview", + "name": "OpenAI o1-preview", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-06-27", - "last_updated": "2024-06-27", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2024-09-12", + "last_updated": "2024-09-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0.2, - "output": 0.2 + "input": 14.993999999999998, + "output": 59.993 } }, - "mistral-saba-24b": { - "id": "mistral-saba-24b", - "name": "Mistral Saba 24B", - "family": "mistral", + "openai/o3-mini": { + "id": "openai/o3-mini", + "name": "OpenAI o3-mini", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-02-06", - "last_updated": "2025-02-06", + "structured_output": true, + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 200000, + "input": 200000, + "output": 100000 }, - "status": "deprecated", "cost": { - "input": 0.79, - "output": 0.79 + "input": 1.1, + "output": 4.4 } }, - "deepseek-r1-distill-llama-70b": { - "id": "deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", - "attachment": false, + "openai/gpt-5.2": { + "id": "openai/gpt-5.2", + "name": "GPT 5.2", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "structured_output": true, + "release_date": "2026-01-01", + "last_updated": "2026-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 400000, + "input": 400000, + "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0.75, - "output": 0.99 + "input": 1.75, + "output": 14 } }, - "llama-guard-3-8b": { - "id": "llama-guard-3-8b", - "name": "Llama Guard 3 8B", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "openai/gpt-5.3-codex": { + "id": "openai/gpt-5.3-codex", + "name": "GPT 5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-24", + "last_updated": "2026-02-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 400000, + "input": 400000, + "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0.2, - "output": 0.2 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "llama-3.3-70b-versatile": { - "id": "llama-3.3-70b-versatile", - "name": "Llama 3.3 70B Versatile", - "family": "llama", - "attachment": false, - "reasoning": false, + "openai/gpt-latest": { + "id": "openai/gpt-latest", + "name": "GPT Latest", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "structured_output": true, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.59, - "output": 0.79 + "input": 5, + "output": 30, + "cache_read": 0.5 + } + }, + "openai/gpt-5.1-codex-mini": { + "id": "openai/gpt-5.1-codex-mini", + "name": "GPT 5.1 Codex Mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 400000, + "input": 400000, + "output": 128000 + }, + "cost": { + "input": 0.25, + "output": 2 } }, - "allam-2-7b": { - "id": "allam-2-7b", - "name": "ALLaM-2-7b", - "family": "allam", + "openai/o4-mini-deep-research": { + "id": "openai/o4-mini-deep-research", + "name": "OpenAI o4-mini Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o-mini", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-09", - "last_updated": "2024-09", + "structured_output": false, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0, - "output": 0 + "input": 9.996, + "output": 19.992 } }, - "whisper-large-v3": { - "id": "whisper-large-v3", - "name": "Whisper Large V3", - "family": "whisper", - "attachment": false, + "openai/gpt-4.1-nano": { + "id": "openai/gpt-4.1-nano", + "name": "GPT 4.1 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2023-09-01", - "last_updated": "2025-09-05", + "structured_output": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { - "input": ["audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 448, - "output": 448 + "context": 1047576, + "input": 1047576, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1, + "output": 0.4 } }, - "llama-3.1-8b-instant": { - "id": "llama-3.1-8b-instant", - "name": "Llama 3.1 8B Instant", - "family": "llama", + "openai/gpt-oss-120b": { + "id": "openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "structured_output": true, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { "input": 0.05, - "output": 0.08 + "output": 0.25 } }, - "llama3-70b-8192": { - "id": "llama3-70b-8192", - "name": "Llama 3 70B", - "family": "llama", - "attachment": false, + "openai/gpt-4o-2024-11-20": { + "id": "openai/gpt-4o-2024-11-20", + "name": "GPT-4o (2024-11-20)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-03", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "tool_call": false, + "structured_output": false, + "release_date": "2024-11-20", + "last_updated": "2024-11-20", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 16384 }, - "status": "deprecated", "cost": { - "input": 0.59, - "output": 0.79 + "input": 2.5, + "output": 10 } }, - "qwen-qwq-32b": { - "id": "qwen-qwq-32b", - "name": "Qwen QwQ 32B", - "family": "qwen", + "openai/o1": { + "id": "openai/o1", + "name": "OpenAI o1", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-11-27", - "last_updated": "2024-11-27", + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-17", + "last_updated": "2024-12-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 200000, + "input": 200000, + "output": 100000 }, - "status": "deprecated", "cost": { - "input": 0.29, - "output": 0.39 + "input": 14.993999999999998, + "output": 59.993 } }, - "whisper-large-v3-turbo": { - "id": "whisper-large-v3-turbo", - "name": "Whisper Large v3 Turbo", - "family": "whisper", - "attachment": false, + "openai/o1-pro": { + "id": "openai/o1-pro", + "name": "OpenAI o1 Pro", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-pro", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "structured_output": false, + "release_date": "2025-01-25", + "last_updated": "2025-01-25", "modalities": { - "input": ["audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 448, - "output": 448 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0, - "output": 0 + "input": 150, + "output": 600 } }, - "llama3-8b-8192": { - "id": "llama3-8b-8192", - "name": "Llama 3 8B", - "family": "llama", - "attachment": false, + "openai/gpt-chat-latest": { + "id": "openai/gpt-chat-latest", + "name": "GPT Chat Latest", + "description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows", + "attachment": true, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-03", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "structured_output": true, + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 400000, + "input": 400000, + "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0.05, - "output": 0.08 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "canopylabs/orpheus-arabic-saudi": { - "id": "canopylabs/orpheus-arabic-saudi", - "name": "Orpheus Arabic Saudi", - "family": "canopylabs", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-12-16", - "release_date": "2025-12-16", - "last_updated": "2025-12-16", + "openai/gpt-5.4": { + "id": "openai/gpt-5.4", + "name": "GPT 5.4", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 4000, - "output": 50000 + "context": 922000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 40, - "output": 0 + "input": 2.5, + "output": 15, + "cache_read": 0.25 } }, - "canopylabs/orpheus-v1-english": { - "id": "canopylabs/orpheus-v1-english", - "name": "Orpheus V1 English", - "family": "canopylabs", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2025-12-19", - "release_date": "2025-12-19", - "last_updated": "2025-12-19", + "openai/gpt-5.4-mini": { + "id": "openai/gpt-5.4-mini", + "name": "GPT 5.4 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { - "input": ["text"], - "output": ["audio"] + "input": ["text", "image", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 4000, - "output": 50000 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "meta-llama/llama-4-scout-17b-16e-instruct": { - "id": "meta-llama/llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B", - "family": "llama", - "attachment": false, + "openai/gpt-4.1": { + "id": "openai/gpt-4.1", + "name": "GPT 4.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "release_date": "2025-09-10", + "last_updated": "2025-09-10", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1047576, + "input": 1047576, + "output": 32768 }, "cost": { - "input": 0.11, - "output": 0.34 + "input": 2, + "output": 8 } }, - "meta-llama/llama-prompt-guard-2-22m": { - "id": "meta-llama/llama-prompt-guard-2-22m", - "name": "Llama Prompt Guard 2 22M", - "family": "llama", + "openai/o3-deep-research": { + "id": "openai/o3-deep-research", + "name": "OpenAI o3 Deep Research", + "description": "Research model for long-horizon investigation, synthesis, and analytical reports", + "family": "o", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["medium"] + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "structured_output": false, + "release_date": "2025-04-16", + "last_updated": "2025-04-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512, - "output": 512 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.03, - "output": 0.03 + "input": 9.996, + "output": 19.992 } }, - "meta-llama/llama-guard-4-12b": { - "id": "meta-llama/llama-guard-4-12b", - "name": "Llama Guard 4 12B", - "family": "llama", + "openai/gpt-4-turbo-preview": { + "id": "openai/gpt-4-turbo-preview", + "name": "GPT-4 Turbo Preview", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "structured_output": false, + "release_date": "2023-11-06", + "last_updated": "2024-01-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 128000, + "output": 4096 + }, + "cost": { + "input": 9.996, + "output": 30.004999999999995 + } + }, + "openai/gpt-5-mini": { + "id": "openai/gpt-5-mini", + "name": "GPT 5 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 1024 + "context": 400000, + "input": 400000, + "output": 128000 }, - "status": "deprecated", "cost": { - "input": 0.2, - "output": 0.2 + "input": 0.25, + "output": 2 } }, - "meta-llama/llama-4-maverick-17b-128e-instruct": { - "id": "meta-llama/llama-4-maverick-17b-128e-instruct", - "name": "Llama 4 Maverick 17B", - "family": "llama", - "attachment": false, + "openai/gpt-4.1-mini": { + "id": "openai/gpt-4.1-mini", + "name": "GPT 4.1 Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1047576, + "input": 1047576, + "output": 32768 }, - "status": "deprecated", "cost": { - "input": 0.2, - "output": 0.6 + "input": 0.4, + "output": 1.6 } }, - "meta-llama/llama-prompt-guard-2-86m": { - "id": "meta-llama/llama-prompt-guard-2-86m", - "name": "Llama Prompt Guard 2 86M", - "family": "llama", - "attachment": false, + "openai/gpt-4-turbo": { + "id": "openai/gpt-4-turbo", + "name": "GPT-4 Turbo", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-10-01", - "last_updated": "2024-10-01", + "structured_output": false, + "release_date": "2023-11-06", + "last_updated": "2024-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512, - "output": 512 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 10, + "output": 30 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, + "openai/gpt-5-nano": { + "id": "openai/gpt-5-nano", + "name": "GPT 5 Nano", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-nano", + "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.05, + "output": 0.4 } }, - "openai/gpt-oss-safeguard-20b": { - "id": "openai/gpt-oss-safeguard-20b", - "name": "Safety GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, + "openai/gpt-5.4-pro": { + "id": "openai/gpt-5.4-pro", + "name": "GPT 5.4 Pro", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-03-05", - "last_updated": "2025-03-05", + "structured_output": true, + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 922000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.075, - "output": 0.3, - "cache_read": 0.037 + "input": 30, + "output": 180, + "cache_read": 3 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "openai/o3-mini-high": { + "id": "openai/o3-mini-high", + "name": "OpenAI o3-mini (High)", + "description": "O-series reasoning model for hard analysis, math, coding, and planning", + "family": "o-mini", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2025-01-31", + "last_updated": "2025-01-31", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 200000, + "input": 200000, + "output": 100000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.64, + "output": 2.588 } }, - "qwen/qwen3-32b": { - "id": "qwen/qwen3-32b", - "name": "Qwen3 32B", - "family": "qwen", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11-08", - "release_date": "2024-12-23", - "last_updated": "2024-12-23", + "openai/gpt-4o-search-preview": { + "id": "openai/gpt-4o-search-preview", + "name": "GPT-4o Search Preview", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-05-13", + "last_updated": "2024-05-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 40960 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.29, - "output": 0.59 + "input": 1.47, + "output": 5.88 } }, - "groq/compound": { - "id": "groq/compound", - "name": "Compound", - "family": "groq", + "openai/gpt-5.1-2025-11-13": { + "id": "openai/gpt-5.1-2025-11-13", + "name": "GPT-5.1 (2025-11-13)", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-09-04", - "release_date": "2025-09-04", - "last_updated": "2025-09-04", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10 } }, - "groq/compound-mini": { - "id": "groq/compound-mini", - "name": "Compound Mini", - "family": "groq", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-09-04", - "release_date": "2025-09-04", - "last_updated": "2025-09-04", + "openai/gpt-4o-mini": { + "id": "openai/gpt-4o-mini", + "name": "GPT-4o mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt-mini", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.1496, + "output": 0.595 } }, - "moonshotai/kimi-k2-instruct": { - "id": "moonshotai/kimi-k2-instruct", - "name": "Kimi K2 Instruct", - "family": "kimi", + "openai/gpt-oss-20b": { + "id": "openai/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-14", - "last_updated": "2025-07-14", + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 128000, + "input": 128000, "output": 16384 }, - "status": "deprecated", "cost": { - "input": 1, - "output": 3 + "input": 0.04, + "output": 0.15 } }, - "moonshotai/kimi-k2-instruct-0905": { - "id": "moonshotai/kimi-k2-instruct-0905", - "name": "Kimi K2 Instruct 0905", - "family": "kimi", + "openai/gpt-5-codex": { + "id": "openai/gpt-5-codex", + "name": "GPT-5 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-15", + "last_updated": "2025-09-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 16384 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 1, - "output": 3 + "input": 9.996, + "output": 19.992 } - } - } - }, - "azure": { - "id": "azure", - "env": ["AZURE_RESOURCE_NAME", "AZURE_API_KEY"], - "npm": "@ai-sdk/azure", - "name": "Azure", - "doc": "https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models", - "models": { - "mistral-nemo": { - "id": "mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, - "reasoning": false, + }, + "openai/gpt-5.2-codex": { + "id": "openai/gpt-5.2-codex", + "name": "GPT 5.2 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "structured_output": true, + "release_date": "2026-01-14", + "last_updated": "2026-01-14", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 400000, + "input": 400000, "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.15 + "input": 1.75, + "output": 14 } }, - "gpt-5.2-chat": { - "id": "gpt-5.2-chat", - "name": "GPT-5.2 Chat", - "family": "gpt-codex", + "openai/gpt-5.1": { + "id": "openai/gpt-5.1", + "name": "GPT 5.1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 400000, + "input": 400000, + "output": 128000 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 1.25, + "output": 10 } }, - "codex-mini": { - "id": "codex-mini", - "name": "Codex Mini", - "family": "gpt-codex-mini", + "openai/gpt-5.5": { + "id": "openai/gpt-5.5", + "name": "GPT 5.5", + "description": "Frontier GPT model for professional reasoning, coding, and multimodal work", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-05-16", - "last_updated": "2025-05-16", + "structured_output": true, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 1.5, - "output": 6, - "cache_read": 0.375 + "input": 5, + "output": 30, + "cache_read": 0.5 } }, - "phi-4-multimodal": { - "id": "phi-4-multimodal", - "name": "Phi-4-multimodal", - "family": "phi", - "attachment": true, + "Gryphe/MythoMax-L2-13b": { + "id": "Gryphe/MythoMax-L2-13b", + "name": "MythoMax 13B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": false, + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 4000, + "input": 4000, "output": 4096 }, "cost": { - "input": 0.08, - "output": 0.32, - "input_audio": 4 + "input": 0.1003, + "output": 0.1003 } }, - "phi-3.5-mini-instruct": { - "id": "phi-3.5-mini-instruct", - "name": "Phi-3.5-mini-instruct", - "family": "phi", + "Unbabel/M-Prometheus-14B": { + "id": "Unbabel/M-Prometheus-14B", + "name": "M-Prometheus 14B", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "structured_output": false, + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 0.2, + "output": 0.2 } }, - "llama-4-scout-17b-16e-instruct": { - "id": "llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", - "attachment": true, + "LLM360/K2-Think": { + "id": "LLM360/K2-Think", + "name": "K2-Think", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "input": 128000, + "output": 32768 }, "cost": { - "input": 0.2, - "output": 0.78 + "input": 0.17, + "output": 0.68 } }, - "grok-4-1-fast-reasoning": { - "id": "grok-4-1-fast-reasoning", - "name": "Grok 4.1 Fast (Reasoning)", - "family": "grok", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-06-27", - "last_updated": "2025-06-27", + "NousResearch/hermes-4-405b": { + "id": "NousResearch/hermes-4-405b", + "name": "Hermes 4 Large", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": true, + "release_date": "2025-08-26", + "last_updated": "2025-08-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, @@ -106565,1522 +160602,1692 @@ "input": 128000, "output": 8192 }, - "status": "beta", "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.3, + "output": 1.2 } }, - "phi-3-medium-4k-instruct": { - "id": "phi-3-medium-4k-instruct", - "name": "Phi-3-medium-instruct (4k)", - "family": "phi", + "NousResearch/hermes-3-llama-3.1-70b": { + "id": "NousResearch/hermes-3-llama-3.1-70b", + "name": "Hermes 3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": false, + "release_date": "2026-01-07", + "last_updated": "2026-01-07", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 1024 + "context": 65536, + "input": 65536, + "output": 8192 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.408, + "output": 0.408 } }, - "ministral-3b": { - "id": "ministral-3b", - "name": "Ministral 3B", - "family": "ministral", + "NousResearch/Hermes-4-70B:thinking": { + "id": "NousResearch/Hermes-4-70B:thinking", + "name": "Hermes 4 (Thinking)", + "description": "General-purpose chat model for instruction following, writing, and analysis", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-17", + "last_updated": "2025-09-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 8192 }, "cost": { - "input": 0.04, - "output": 0.04 + "input": 0.2006, + "output": 0.3995 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-02-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "NousResearch/hermes-4-70b": { + "id": "NousResearch/hermes-4-70b", + "name": "Hermes 4 Medium", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-03", + "last_updated": "2025-07-03", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.2006, + "output": 0.3995 } }, - "meta-llama-3.1-8b-instruct": { - "id": "meta-llama-3.1-8b-instruct", - "name": "Meta-Llama-3.1-8B-Instruct", - "family": "llama", + "NousResearch/DeepHermes-3-Mistral-24B-Preview": { + "id": "NousResearch/DeepHermes-3-Mistral-24B-Preview", + "name": "DeepHermes-3 Mistral 24B (Preview)", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-10", + "last_updated": "2025-05-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 32768 }, "cost": { "input": 0.3, - "output": 0.61 + "output": 0.3 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", + "NousResearch/hermes-4-405b:thinking": { + "id": "NousResearch/hermes-4-405b:thinking", + "name": "Hermes 4 Large (Thinking)", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": true, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-06", - "last_updated": "2026-02-06", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", - "shape": "completions" + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 3 + "input": 0.3, + "output": 1.2 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", - "attachment": false, + "unsloth/gemma-3-12b-it": { + "id": "unsloth/gemma-3-12b-it", + "name": "Gemma 3 12B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "unsloth", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 32768 + "input": 128000, + "output": 131072 }, "cost": { - "input": 0.71, - "output": 0.71 + "input": 0.272, + "output": 0.272 } }, - "deepseek-v3-0324": { - "id": "deepseek-v3-0324", - "name": "DeepSeek-V3-0324", - "family": "deepseek", - "attachment": false, + "unsloth/gemma-3-4b-it": { + "id": "unsloth/gemma-3-4b-it", + "name": "Gemma 3 4B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "unsloth", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 1.14, - "output": 4.56 + "input": 0.2006, + "output": 0.2006 } }, - "gpt-5-chat": { - "id": "gpt-5-chat", - "name": "GPT-5 Chat", - "family": "gpt-codex", + "unsloth/gemma-3-27b-it": { + "id": "unsloth/gemma-3-27b-it", + "name": "Gemma 3 27B IT", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "unsloth", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2024-10-24", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": false, + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "input": 128000, + "output": 96000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.2992, + "output": 0.2992 } }, - "phi-3.5-moe-instruct": { - "id": "phi-3.5-moe-instruct", - "name": "Phi-3.5-MoE-instruct", - "family": "phi", + "NeverSleep/Lumimaid-v0.2-70B": { + "id": "NeverSleep/Lumimaid-v0.2-70B", + "name": "Lumimaid v0.2", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.16, - "output": 0.64 + "input": 1, + "output": 1.5 } }, - "gpt-5.3-chat": { - "id": "gpt-5.3-chat", - "name": "GPT-5.3 Chat", - "family": "gpt-codex", + "mistralai/mixtral-8x7b-instruct-v0.1": { + "id": "mistralai/mixtral-8x7b-instruct-v0.1", + "name": "Mixtral 8x7B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mixtral", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "input": 32768, + "output": 32768 + }, + "cost": { + "input": 0.27, + "output": 0.27 + } + }, + "mistralai/mistral-small-4-119b-2603:thinking": { + "id": "mistralai/mistral-small-4-119b-2603:thinking", + "name": "Mistral Small 4 119B Thinking", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 262144, + "input": 262144, "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.4, + "output": 1.4 } }, - "o1-mini": { - "id": "o1-mini", - "name": "o1-mini", - "family": "o-mini", + "mistralai/mixtral-8x22b-instruct-v0.1": { + "id": "mistralai/mixtral-8x22b-instruct-v0.1", + "name": "Mixtral 8x22B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mixtral", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 65536, + "input": 65536, + "output": 32768 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.8999999999999999, + "output": 0.8999999999999999 } }, - "text-embedding-3-large": { - "id": "text-embedding-3-large", - "name": "text-embedding-3-large", - "family": "text-embedding", + "mistralai/Devstral-Small-2505": { + "id": "mistralai/Devstral-Small-2505", + "name": "Mistral Devstral Small 2505", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "structured_output": false, + "release_date": "2025-08-02", + "last_updated": "2025-08-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 3072 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 0.13, - "output": 0 + "input": 0.060000000000000005, + "output": 0.060000000000000005 } }, - "phi-3-mini-128k-instruct": { - "id": "phi-3-mini-128k-instruct", - "name": "Phi-3-mini-instruct (128k)", - "family": "phi", + "mistralai/ministral-8b-2512": { + "id": "mistralai/ministral-8b-2512", + "name": "Ministral 8B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": false, + "release_date": "2025-12-04", + "last_updated": "2025-12-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 32768 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 0.15, + "output": 0.15 } }, - "phi-4-reasoning": { - "id": "phi-4-reasoning", - "name": "Phi-4-reasoning", - "family": "phi", + "mistralai/mistral-saba": { + "id": "mistralai/mistral-saba", + "name": "Mistral Saba", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": false, + "release_date": "2025-02-17", + "last_updated": "2025-02-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 32000, - "output": 4096 + "input": 32000, + "output": 32768 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 0.1989, + "output": 0.595 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "mistralai/mistral-large": { + "id": "mistralai/mistral-large", + "name": "Mistral Large 2411", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-02-26", + "last_updated": "2024-02-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 256000 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.03 + "input": 2.006, + "output": 6.001 } }, - "gpt-5-nano": { - "id": "gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "mistralai/mistral-medium-3.1": { + "id": "mistralai/mistral-medium-3.1", + "name": "Mistral Medium 3.1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 131072, + "input": 131072, + "output": 32768 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.01 + "input": 0.4, + "output": 2 } }, - "meta-llama-3-70b-instruct": { - "id": "meta-llama-3-70b-instruct", - "name": "Meta-Llama-3-70B-Instruct", - "family": "llama", + "mistralai/ministral-3b-2512": { + "id": "mistralai/ministral-3b-2512", + "name": "Ministral 3B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "structured_output": false, + "release_date": "2025-12-04", + "last_updated": "2025-12-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 131072, + "input": 131072, + "output": 32768 }, "cost": { - "input": 2.68, - "output": 3.54 + "input": 0.1, + "output": 0.1 } }, - "phi-3-small-8k-instruct": { - "id": "phi-3-small-8k-instruct", - "name": "Phi-3-small-instruct (8k)", - "family": "phi", - "attachment": false, + "mistralai/ministral-14b-instruct-2512": { + "id": "mistralai/ministral-14b-instruct-2512", + "name": "Ministral 3 14B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", + "attachment": true, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": false, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 262144, + "input": 262144, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.1, + "output": 0.4 } }, - "gpt-5.3-codex": { - "id": "gpt-5.3-codex", - "name": "GPT-5.3 Codex", - "family": "gpt-codex", - "attachment": false, + "mistralai/mistral-small-4-119b-2603": { + "id": "mistralai/mistral-small-4-119b-2603", + "name": "Mistral Small 4 119B", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-02-24", - "last_updated": "2026-02-24", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.4, + "output": 1.4 } }, - "text-embedding-ada-002": { - "id": "text-embedding-ada-002", - "name": "text-embedding-ada-002", - "family": "text-embedding", + "mistralai/ministral-14b-2512": { + "id": "mistralai/ministral-14b-2512", + "name": "Ministral 14B", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", + "family": "ministral", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2022-12-15", - "last_updated": "2022-12-15", + "structured_output": false, + "release_date": "2025-12-04", + "last_updated": "2025-12-04", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 1536 + "context": 262144, + "input": 262144, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.2, + "output": 0.2 } }, - "llama-3.2-90b-vision-instruct": { - "id": "llama-3.2-90b-vision-instruct", - "name": "Llama-3.2-90B-Vision-Instruct", - "family": "llama", + "mistralai/mistral-large-3-675b-instruct-2512": { + "id": "mistralai/mistral-large-3-675b-instruct-2512", + "name": "Mistral Large 3 675B", + "description": "Flagship Mistral model for advanced reasoning, coding, and multilingual work", + "family": "mistral-large", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 256000 }, "cost": { - "input": 2.04, - "output": 2.04 + "input": 1, + "output": 3 } }, - "deepseek-r1": { - "id": "deepseek-r1", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", + "mistralai/mistral-medium-3": { + "id": "mistralai/mistral-medium-3", + "name": "Mistral Medium 3", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-medium", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 131072, + "input": 131072, + "output": 32768 + }, + "cost": { + "input": 0.4, + "output": 2 + } + }, + "mistralai/devstral-2-123b-instruct-2512": { + "id": "mistralai/devstral-2-123b-instruct-2512", + "name": "Devstral 2 123B", + "description": "Mistral coding agent model for repository tasks and software engineering workflows", + "family": "devstral", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "structured_output": false, + "release_date": "2025-12-09", + "last_updated": "2025-12-09", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 262144, + "input": 262144, + "output": 65536 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.4, + "output": 1.4 } }, - "grok-4-1-fast-non-reasoning": { - "id": "grok-4-1-fast-non-reasoning", - "name": "Grok 4.1 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, + "mistralai/codestral-2508": { + "id": "mistralai/codestral-2508", + "name": "Codestral 2508", + "description": "Mistral coding model for code completion, generation, and developer workflows", + "family": "codestral", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2025-06-27", - "last_updated": "2025-06-27", + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-01", + "last_updated": "2025-08-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "input": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 32768 }, - "status": "beta", "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.3, + "output": 0.8999999999999999 } }, - "deepseek-v3.2-speciale": { - "id": "deepseek-v3.2-speciale", - "name": "DeepSeek-V3.2-Speciale", - "family": "deepseek", + "mistralai/Mistral-Nemo-Instruct-2407": { + "id": "mistralai/Mistral-Nemo-Instruct-2407", + "name": "Mistral Nemo", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "structured_output": false, + "release_date": "2024-07-18", + "last_updated": "2024-07-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.58, - "output": 1.68 + "input": 0.1003, + "output": 0.1207 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, + "bytedance-seed/seed-2.0-lite": { + "id": "bytedance-seed/seed-2.0-lite", + "name": "ByteDance Seed 2.0 Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "attachment": false, + "reasoning": false, + "tool_call": false, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "release_date": "2026-03-10", + "last_updated": "2026-03-10", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.125 + "input": 0.25, + "output": 2 } }, - "mistral-large-2411": { - "id": "mistral-large-2411", - "name": "Mistral Large 24.11", - "family": "mistral-large", + "anthracite-org/magnum-v2-72b": { + "id": "anthracite-org/magnum-v2-72b", + "name": "Magnum V2 72B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 2, - "output": 6 + "input": 2.006, + "output": 2.992 } }, - "claude-opus-4-1": { - "id": "claude-opus-4-1", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "anthracite-org/magnum-v4-72b": { + "id": "anthracite-org/magnum-v4-72b", + "name": "Magnum v4 72B", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 2.006, + "output": 2.992 } }, - "cohere-command-a": { - "id": "cohere-command-a", - "name": "Command A", - "family": "command-a", + "inflatebot/MN-12B-Mag-Mell-R1": { + "id": "inflatebot/MN-12B-Mag-Mell-R1", + "name": "Mag Mell R1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8000 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "llama-3.2-11b-vision-instruct": { - "id": "llama-3.2-11b-vision-instruct", - "name": "Llama-3.2-11B-Vision-Instruct", - "family": "llama", + "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": { + "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning", + "name": "Nvidia Nemotron 3 Nano Omni", + "description": "Open Nemotron omni model combining reasoning with text, vision, and audio", + "family": "nemotron", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "release_date": "2026-04-28", + "last_updated": "2026-04-28", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 65536 }, "cost": { - "input": 0.37, - "output": 0.37 + "input": 0.105, + "output": 0.42 } }, - "meta-llama-3.1-405b-instruct": { - "id": "meta-llama-3.1-405b-instruct", - "name": "Meta-Llama-3.1-405B-Instruct", - "family": "llama", + "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF": { + "id": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "name": "Nvidia Nemotron 70b", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 5.33, - "output": 16 + "input": 0.357, + "output": 0.408 } }, - "gpt-5.1-chat": { - "id": "gpt-5.1-chat", - "name": "GPT-5.1 Chat", - "family": "gpt-codex", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "nvidia/nvidia-nemotron-nano-9b-v2": { + "id": "nvidia/nvidia-nemotron-nano-9b-v2", + "name": "Nvidia Nemotron Nano 9B v2", + "description": "Compact Nemotron model for efficient reasoning and deployable AI agents", + "family": "nemotron", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-08-18", + "last_updated": "2025-08-18", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, + "input": 128000, "output": 16384 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.17, + "output": 0.68 } }, - "gpt-4-turbo-vision": { - "id": "gpt-4-turbo-vision", - "name": "GPT-4 Turbo Vision", - "family": "gpt", - "attachment": true, + "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5": { + "id": "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5", + "name": "Nvidia Nemotron Super 49B v1.5", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", + "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, + "structured_output": false, "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "input": 128000, + "output": 16384 }, "cost": { - "input": 10, - "output": 30 + "input": 0.05, + "output": 0.25 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", + "nvidia/nemotron-3-nano-30b-a3b": { + "id": "nvidia/nemotron-3-nano-30b-a3b", + "name": "Nvidia Nemotron 3 Nano 30B", + "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents", + "family": "nemotron", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-01-14", - "last_updated": "2026-01-14", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "temperature": true, + "release_date": "2025-12-15", + "last_updated": "2025-12-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 1.75, - "output": 14, - "cache_read": 0.175 + "input": 0.17, + "output": 0.68 } }, - "cohere-embed-v-4-0": { - "id": "cohere-embed-v-4-0", - "name": "Embed v4", - "family": "cohere-embed", - "attachment": true, + "nvidia/Llama-3.3-Nemotron-Super-49B-v1": { + "id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1", + "name": "Nvidia Nemotron Super 49B", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "release_date": "2025-04-15", - "last_updated": "2025-04-15", + "structured_output": false, + "temperature": true, + "release_date": "2025-08-08", + "last_updated": "2025-08-08", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 1536 + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.12, - "output": 0 + "input": 0.15, + "output": 0.15 } }, - "gpt-5.1-codex-mini": { - "id": "gpt-5.1-codex-mini", - "name": "GPT-5.1 Codex Mini", - "family": "gpt-codex", + "nvidia/nemotron-3-super-120b-a12b:thinking": { + "id": "nvidia/nemotron-3-super-120b-a12b:thinking", + "name": "Nvidia Nemotron 3 Super 120B Thinking", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "structured_output": false, + "temperature": true, + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.05, + "output": 0.25 } }, - "gpt-3.5-turbo-0125": { - "id": "gpt-3.5-turbo-0125", - "name": "GPT-3.5 Turbo 0125", - "family": "gpt", + "nvidia/nemotron-3-super-120b-a12b": { + "id": "nvidia/nemotron-3-super-120b-a12b", + "name": "Nvidia Nemotron 3 Super 120B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": false, "temperature": true, - "knowledge": "2021-08", - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, + "context": 262144, + "input": 262144, "output": 16384 }, "cost": { - "input": 0.5, - "output": 1.5 + "input": 0.05, + "output": 0.25 } }, - "o1-preview": { - "id": "o1-preview", - "name": "o1-preview", - "family": "o", + "cognitivecomputations/dolphin-2.9.2-qwen2-72b": { + "id": "cognitivecomputations/dolphin-2.9.2-qwen2-72b", + "name": "Dolphin 72b", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 8192, + "input": 8192, + "output": 4096 }, "cost": { - "input": 16.5, - "output": 66, - "cache_read": 8.25 + "input": 0.306, + "output": 0.306 } }, - "cohere-embed-v3-multilingual": { - "id": "cohere-embed-v3-multilingual", - "name": "Embed v3 Multilingual", - "family": "cohere-embed", + "xiaomi/mimo-v2-flash-original": { + "id": "xiaomi/mimo-v2-flash-original", + "name": "MiMo V2 Flash Original", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "temperature": false, - "release_date": "2023-11-07", - "last_updated": "2023-11-07", + "structured_output": false, + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512, - "output": 1024 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.102, + "output": 0.306 } }, - "grok-4-20-non-reasoning": { - "id": "grok-4-20-non-reasoning", - "name": "Grok 4.20 (Non-Reasoning)", - "family": "grok", + "xiaomi/mimo-v2-flash-thinking-original": { + "id": "xiaomi/mimo-v2-flash-thinking-original", + "name": "MiMo V2 Flash (Thinking) Original", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-09", - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 32768 }, - "status": "beta", "cost": { - "input": 2, - "output": 6 + "input": 0.102, + "output": 0.306 } }, - "gpt-5.1": { - "id": "gpt-5.1", - "name": "GPT-5.1", - "family": "gpt", + "xiaomi/mimo-v2.5": { + "id": "xiaomi/mimo-v2.5", + "name": "MiMo V2.5", + "description": "MiMo omni model for text, image, video, audio, and agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] + "input": ["text", "image", "video"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 1048576, + "input": 1048576, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "grok-4-fast-reasoning": { - "id": "grok-4-fast-reasoning", - "name": "Grok 4 Fast (Reasoning)", - "family": "grok", + "xiaomi/mimo-v2-omni": { + "id": "xiaomi/mimo-v2-omni", + "name": "MiMo V2 Omni", + "description": "MiMo omni model for text, image, video, audio, and agents", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "structured_output": false, + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 262144, + "input": 262144, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.4, + "output": 2, + "cache_read": 0.08 } }, - "o1": { - "id": "o1", - "name": "o1", - "family": "o", + "xiaomi/mimo-v2-flash": { + "id": "xiaomi/mimo-v2-flash", + "name": "MiMo V2 Flash", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2023-09", - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 15, - "output": 60, - "cache_read": 7.5 + "input": 0.102, + "output": 0.306 } }, - "mistral-small-2503": { - "id": "mistral-small-2503", - "name": "Mistral Small 3.1", - "family": "mistral-small", - "attachment": true, - "reasoning": false, + "xiaomi/mimo-v2-pro": { + "id": "xiaomi/mimo-v2-pro", + "name": "MiMo V2 Pro", + "description": "MiMo pro model for strong multimodal reasoning and agent execution", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "structured_output": true, + "release_date": "2026-03-19", + "last_updated": "2026-03-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1048576, + "input": 1048576, + "output": 131072 }, "cost": { - "input": 0.1, - "output": 0.3 + "input": 1, + "output": 3, + "cache_read": 0.2 } }, - "model-router": { - "id": "model-router", - "name": "Model Router", - "family": "model-router", - "attachment": true, - "reasoning": false, + "xiaomi/mimo-v2.5-pro": { + "id": "xiaomi/mimo-v2.5-pro", + "name": "MiMo V2.5 Pro", + "description": "MiMo pro model for strong multimodal reasoning and agent execution", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "release_date": "2025-05-19", - "last_updated": "2025-11-18", + "structured_output": true, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "input": 1048576, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0 + "input": 0.435, + "output": 0.87, + "cache_read": 0.0036 } }, - "gpt-3.5-turbo-1106": { - "id": "gpt-3.5-turbo-1106", - "name": "GPT-3.5 Turbo 1106", - "family": "gpt", + "xiaomi/mimo-v2-flash-thinking": { + "id": "xiaomi/mimo-v2-flash-thinking", + "name": "MiMo V2 Flash (Thinking)", + "description": "MiMo flash model for fast multimodal assistance and agent workflows", + "family": "mimo", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": false, - "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-11-06", - "last_updated": "2023-11-06", + "structured_output": false, + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 1, - "output": 2 + "input": 0.102, + "output": 0.306 } }, - "text-embedding-3-small": { - "id": "text-embedding-3-small", - "name": "text-embedding-3-small", - "family": "text-embedding", - "attachment": false, - "reasoning": false, - "tool_call": false, - "release_date": "2024-01-25", - "last_updated": "2024-01-25", + "anthropic/claude-haiku-latest": { + "id": "anthropic/claude-haiku-latest", + "name": "Claude Haiku Latest", + "description": "Fast Claude model for responsive assistance, classification, and lightweight agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8191, - "output": 1536 + "context": 200000, + "input": 200000, + "output": 64000 }, "cost": { - "input": 0.02, - "output": 0 + "input": 1, + "output": 5, + "cache_read": 0.1 } }, - "deepseek-v3.1": { - "id": "deepseek-v3.1", - "name": "DeepSeek-V3.1", - "family": "deepseek", - "attachment": false, + "anthropic/claude-opus-4.7:thinking": { + "id": "anthropic/claude-opus-4.7:thinking", + "name": "Claude 4.7 Opus Thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "structured_output": true, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.56, - "output": 1.68 + "input": 4.998, + "output": 25.007, + "cache_read": 0.4998 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "Claude Opus 4.5", + "anthropic/claude-opus-4.6:thinking:max": { + "id": "anthropic/claude-opus-4.6:thinking:max", + "name": "Claude 4.6 Opus Thinking Max", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-08-01", + "structured_output": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 4.998, + "output": 25.007 } }, - "phi-3-mini-4k-instruct": { - "id": "phi-3-mini-4k-instruct", - "name": "Phi-3-mini-instruct (4k)", - "family": "phi", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "anthropic/claude-opus-4.6:thinking:low": { + "id": "anthropic/claude-opus-4.6:thinking:low", + "name": "Claude 4.6 Opus Thinking Low", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 1024 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.13, - "output": 0.52 + "input": 4.998, + "output": 25.007 } }, - "meta-llama-3.1-70b-instruct": { - "id": "meta-llama-3.1-70b-instruct", - "name": "Meta-Llama-3.1-70B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, + "anthropic/claude-sonnet-4.6:thinking": { + "id": "anthropic/claude-sonnet-4.6:thinking", + "name": "Claude Sonnet 4.6 Thinking", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "structured_output": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 2.68, - "output": 3.54 + "input": 2.992, + "output": 14.993999999999998 } }, - "phi-4-mini-reasoning": { - "id": "phi-4-mini-reasoning", - "name": "Phi-4-mini-reasoning", - "family": "phi", - "attachment": false, + "anthropic/claude-opus-4.7": { + "id": "anthropic/claude-opus-4.7", + "name": "Claude 4.7 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": true, + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 4.998, + "output": 25.007, + "cache_read": 0.4998 } }, - "gpt-4": { - "id": "gpt-4", - "name": "GPT-4", - "family": "gpt", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-4.6:thinking:medium": { + "id": "anthropic/claude-opus-4.6:thinking:medium", + "name": "Claude 4.6 Opus Thinking Medium", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-03-14", - "last_updated": "2023-03-14", + "structured_output": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 8192, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 60, - "output": 120 + "input": 4.998, + "output": 25.007 } }, - "meta-llama-3-8b-instruct": { - "id": "meta-llama-3-8b-instruct", - "name": "Meta-Llama-3-8B-Instruct", - "family": "llama", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "anthropic/claude-opus-4.8:thinking": { + "id": "anthropic/claude-opus-4.8:thinking", + "name": "Claude Opus 4.8 Thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 0.61 + "input": 4.998, + "output": 25.007, + "cache_read": 0.4998 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi", - "attachment": false, + "anthropic/claude-sonnet-latest": { + "id": "anthropic/claude-sonnet-latest", + "name": "Claude Sonnet Latest", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "interleaved": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "release_date": "2026-03-01", + "last_updated": "2026-03-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 - }, - "provider": { - "npm": "@ai-sdk/openai-compatible", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models", - "shape": "completions" + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.95, - "output": 4 + "input": 2.992, + "output": 14.994, + "cache_read": 0.2992 } }, - "gpt-5-codex": { - "id": "gpt-5-codex", - "name": "GPT-5-Codex", - "family": "gpt-codex", - "attachment": false, + "anthropic/claude-opus-4.8": { + "id": "anthropic/claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "structured_output": true, + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, + "context": 1000000, + "input": 1000000, "output": 128000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 4.998, + "output": 25.007, + "cache_read": 0.4998 } }, - "phi-4-mini": { - "id": "phi-4-mini", - "name": "Phi-4-mini", - "family": "phi", - "attachment": false, - "reasoning": false, + "anthropic/claude-opus-latest": { + "id": "anthropic/claude-opus-latest", + "name": "Claude Opus Latest", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": true, + "release_date": "2026-03-29", + "last_updated": "2026-03-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 4.998, + "output": 25.007, + "cache_read": 0.4998 } }, - "grok-4-20-reasoning": { - "id": "grok-4-20-reasoning", - "name": "Grok 4.20 (Reasoning)", - "family": "grok", - "attachment": false, - "reasoning": true, + "anthropic/claude-sonnet-4.6": { + "id": "anthropic/claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-09", - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "structured_output": true, + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-02-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262000, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 128000 }, - "status": "beta", "cost": { - "input": 2, - "output": 6 + "input": 2.992, + "output": 14.993999999999998 } }, - "gpt-3.5-turbo-0301": { - "id": "gpt-3.5-turbo-0301", - "name": "GPT-3.5 Turbo 0301", - "family": "gpt", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-03-01", - "last_updated": "2023-03-01", + "anthropic/claude-opus-4.6:thinking": { + "id": "anthropic/claude-opus-4.6:thinking", + "name": "Claude 4.6 Opus Thinking", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 1000000, + "input": 1000000, + "output": 128000 }, "cost": { - "input": 1.5, - "output": 2 + "input": 4.998, + "output": 25.007 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", + "anthropic/claude-opus-4.6": { + "id": "anthropic/claude-opus-4.6", + "name": "Claude 4.6 Opus", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", + "structured_output": true, "release_date": "2026-02-05", "last_updated": "2026-02-05", "modalities": { @@ -108089,1980 +162296,1969 @@ }, "open_weights": false, "limit": { - "context": 200000, + "context": 1000000, + "input": 1000000, "output": 128000 }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "cost": { + "input": 4.998, + "output": 25.007 + } + }, + "tencent/Hunyuan-MT-7B": { + "id": "tencent/Hunyuan-MT-7B", + "name": "Hunyuan MT 7B", + "description": "Translation model for multilingual conversion, localization, and cross-language workflows", + "family": "hunyuan", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-18", + "last_updated": "2025-09-18", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25, - "tiers": [ - { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 10, - "output": 37.5, - "cache_read": 1, - "cache_write": 12.5 - } + "input": 10, + "output": 20 } }, - "phi-3-small-128k-instruct": { - "id": "phi-3-small-128k-instruct", - "name": "Phi-3-small-instruct (128k)", - "family": "phi", + "tencent/hy3-preview": { + "id": "tencent/hy3-preview", + "name": "Tencent: Hy3 preview", + "description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": false, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 262144 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.066, + "output": 0.26, + "cache_read": 0.029 } }, - "deepseek-v3.2": { - "id": "deepseek-v3.2", - "name": "DeepSeek-V3.2", - "family": "deepseek", + "chutesai/Mistral-Small-3.2-24B-Instruct-2506": { + "id": "chutesai/Mistral-Small-3.2-24B-Instruct-2506", + "name": "Mistral Small 3.2 24b Instruct", + "description": "Efficient Mistral model for fast chat, extraction, and production assistants", + "family": "chutesai", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 128000 + "input": 128000, + "output": 131072 }, "cost": { - "input": 0.58, - "output": 1.68 + "input": 0.2, + "output": 0.4 } }, - "phi-3-medium-128k-instruct": { - "id": "phi-3-medium-128k-instruct", - "name": "Phi-3-medium-instruct (128k)", - "family": "phi", + "dmind/dmind-1-mini": { + "id": "dmind/dmind-1-mini", + "name": "DMind-1-Mini", + "description": "Compact GPT model for low-latency assistance and high-volume workloads", + "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": false, + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 0.17, - "output": 0.68 + "input": 0.2, + "output": 0.4 } }, - "gpt-3.5-turbo-0613": { - "id": "gpt-3.5-turbo-0613", - "name": "GPT-3.5 Turbo 0613", + "dmind/dmind-1": { + "id": "dmind/dmind-1", + "name": "DMind-1", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", "family": "gpt", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-06-13", - "last_updated": "2023-06-13", + "structured_output": false, + "release_date": "2025-06-01", + "last_updated": "2025-06-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 16384, - "output": 16384 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 3, - "output": 4 + "input": 0.3, + "output": 0.6 } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", + "MarinaraSpaghetti/NemoMix-Unleashed-12B": { + "id": "MarinaraSpaghetti/NemoMix-Unleashed-12B", + "name": "NemoMix 12B Unleashed", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "phi-4": { - "id": "phi-4", - "name": "Phi-4", - "family": "phi", + "deepcogito/cogito-v1-preview-qwen-32B": { + "id": "deepcogito/cogito-v1-preview-qwen-32B", + "name": "Cogito v1 Preview Qwen 32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": false, + "release_date": "2025-05-10", + "last_updated": "2025-05-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "input": 128000, + "output": 32768 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 1.7999999999999998, + "output": 1.7999999999999998 } }, - "gpt-5": { - "id": "gpt-5", - "name": "GPT-5", - "family": "gpt", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "cohere/command-r": { + "id": "cohere/command-r", + "name": "Cohere: Command R", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-03-11", + "last_updated": "2024-03-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 272000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.13 + "input": 0.476, + "output": 1.428 } }, - "gpt-4-32k": { - "id": "gpt-4-32k", - "name": "GPT-4 32K", - "family": "gpt", + "cohere/command-r-plus-08-2024": { + "id": "cohere/command-r-plus-08-2024", + "name": "Cohere: Command R+", + "description": "Cohere retrieval model for long-context chat and enterprise RAG workflows", + "family": "command-r", "attachment": false, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-11", - "release_date": "2023-03-14", - "last_updated": "2023-03-14", + "structured_output": false, + "release_date": "2024-08-30", + "last_updated": "2024-08-30", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 128000, + "input": 128000, + "output": 4096 }, "cost": { - "input": 60, - "output": 120 + "input": 2.856, + "output": 14.246 } }, - "cohere-embed-v3-english": { - "id": "cohere-embed-v3-english", - "name": "Embed v3 English", - "family": "cohere-embed", - "attachment": false, - "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2023-11-07", - "last_updated": "2023-11-07", + "stepfun/step-3.7-flash:thinking": { + "id": "stepfun/step-3.7-flash:thinking", + "name": "Step 3.7 Flash Thinking", + "description": "StepFun flash model for efficient multimodal reasoning, coding, and tool use", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "structured_output": true, + "release_date": "2026-05-29", + "last_updated": "2026-05-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512, - "output": 1024 + "context": 256000, + "input": 256000, + "output": 256000 }, "cost": { - "input": 0.1, - "output": 0 + "input": 0.2, + "output": 1.15, + "cache_read": 0.04 } }, - "phi-4-reasoning-plus": { - "id": "phi-4-reasoning-plus", - "name": "Phi-4-reasoning-plus", - "family": "phi", + "nex-agi/deepseek-v3.1-nex-n1": { + "id": "nex-agi/deepseek-v3.1-nex-n1", + "name": "DeepSeek V3.1 Nex N1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": false, + "release_date": "2025-12-10", + "last_updated": "2025-12-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 4096 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.125, - "output": 0.5 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "mistral-medium-2505": { - "id": "mistral-medium-2505", - "name": "Mistral Medium 3", - "family": "mistral-medium", + "undi95/remm-slerp-l2-13b": { + "id": "undi95/remm-slerp-l2-13b", + "name": "ReMM SLERP 13B", + "description": "Open Llama multimodal model for image understanding and text reasoning", + "family": "llama", "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-05", - "release_date": "2025-05-07", - "last_updated": "2025-05-07", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 6144, + "input": 6144, + "output": 4096 }, "cost": { - "input": 0.4, - "output": 2 + "input": 0.7989999999999999, + "output": 1.2069999999999999 } }, - "gpt-3.5-turbo-instruct": { - "id": "gpt-3.5-turbo-instruct", - "name": "GPT-3.5 Turbo Instruct", - "family": "gpt", + "nothingiisreal/L3.1-70B-Celeste-V0.1-BF16": { + "id": "nothingiisreal/L3.1-70B-Celeste-V0.1-BF16", + "name": "Llama 3.1 70B Celeste v0.1", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2021-08", - "release_date": "2023-09-21", - "last_updated": "2023-09-21", + "structured_output": false, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 1.5, - "output": 2 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "deepseek-r1-0528": { - "id": "deepseek-r1-0528", - "name": "DeepSeek-R1-0528", - "family": "deepseek-thinking", + "zai-org/glm-4.7-flash-original:thinking": { + "id": "zai-org/glm-4.7-flash-original:thinking", + "name": "GLM 4.7 Flash Original Thinking", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163840, - "output": 163840 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 0.07, + "output": 0.4 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "zai-org/glm-4.7": { + "id": "zai-org/glm-4.7", + "name": "GLM 4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-12-02", + "structured_output": true, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.15, + "output": 0.8 } }, - "gpt-5.1-codex": { - "id": "gpt-5.1-codex", - "name": "GPT-5.1 Codex", - "family": "gpt-codex", + "zai-org/GLM-4.5-Air:thinking": { + "id": "zai-org/GLM-4.5-Air:thinking", + "name": "GLM 4.5 Air (Thinking)", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-14", - "last_updated": "2025-11-14", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "audio"], - "output": ["text", "image", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 98304 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.12, + "output": 0.8 } }, - "codestral-2501": { - "id": "codestral-2501", - "name": "Codestral 25.01", - "family": "codestral", + "zai-org/GLM-4.5:thinking": { + "id": "zai-org/GLM-4.5:thinking", + "name": "GLM 4.5 (Thinking)", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 256000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { "input": 0.3, - "output": 0.9 + "output": 1.3 } }, - "llama-4-maverick-17b-128e-instruct-fp8": { - "id": "llama-4-maverick-17b-128e-instruct-fp8", - "name": "Llama 4 Maverick 17B 128E Instruct FP8", - "family": "llama", - "attachment": true, + "zai-org/glm-4.5": { + "id": "zai-org/glm-4.5", + "name": "GLM 4.5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-04-05", - "last_updated": "2025-04-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 1 + "input": 0.3, + "output": 1.3 } }, - "mai-ds-r1": { - "id": "mai-ds-r1", - "name": "MAI-DS-R1", - "family": "mai", + "zai-org/glm-5-original": { + "id": "zai-org/glm-5-original", + "name": "GLM 5 Original", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, "reasoning": true, - "tool_call": false, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 1.35, - "output": 5.4 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "gpt-5.1-codex-max": { - "id": "gpt-5.1-codex-max", - "name": "GPT-5.1 Codex Max", - "family": "gpt-codex", - "attachment": true, + "zai-org/glm-5.1": { + "id": "zai-org/glm-5.1", + "name": "GLM 5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-11-13", - "last_updated": "2025-11-13", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 128000 + "context": 200000, + "input": 200000, + "output": 131072 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.3, + "output": 2.55 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "zai-org/glm-4.7-original": { + "id": "zai-org/glm-4.7-original", + "name": "GLM 4.7 Original", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 - }, - "provider": { - "npm": "@ai-sdk/anthropic", - "api": "https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1" + "context": 200000, + "input": 200000, + "output": 65535 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "gpt-5.5": { - "id": "gpt-5.5", - "name": "GPT-5.5", - "family": "gpt", - "attachment": true, + "zai-org/glm-4.7-original:thinking": { + "id": "zai-org/glm-4.7-original:thinking", + "name": "GLM 4.7 Original Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-12-01", - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 200000, + "input": 200000, + "output": 65535 }, "cost": { - "input": 5, - "output": 30, - "cache_read": 0.5, - "context_over_200k": { - "input": 10, - "output": 45, - "cache_read": 1 - }, - "tiers": [ - { - "input": 10, - "output": 45, - "cache_read": 1, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.6, + "output": 2.2, + "cache_read": 0.11 } }, - "gpt-4-turbo": { - "id": "gpt-4-turbo", - "name": "GPT-4 Turbo", - "family": "gpt", - "attachment": true, - "reasoning": false, + "zai-org/glm-4.7-flash-original": { + "id": "zai-org/glm-4.7-flash-original", + "name": "GLM 4.7 Flash Original", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2023-11-06", - "last_updated": "2024-04-09", + "structured_output": true, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 10, - "output": 30 + "input": 0.07, + "output": 0.4 } }, - "gpt-4o-mini": { - "id": "gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": false, + "zai-org/glm-4.7:thinking": { + "id": "zai-org/glm-4.7:thinking", + "name": "GLM 4.7 Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "structured_output": true, + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 200000, + "input": 200000, + "output": 65535 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.08 + "input": 0.2, + "output": 0.8 } }, - "gpt-5.4-mini": { - "id": "gpt-5.4-mini", - "name": "GPT-5.4 Mini", - "family": "gpt-mini", - "attachment": true, + "zai-org/glm-5.1:thinking": { + "id": "zai-org/glm-5.1:thinking", + "name": "GLM 5.1 Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "input": 200000, + "output": 131072 }, "cost": { - "input": 0.75, - "output": 4.5, - "cache_read": 0.075 + "input": 0.3, + "output": 2.55 } }, - "cohere-command-r-08-2024": { - "id": "cohere-command-r-08-2024", - "name": "Command R", - "family": "command-r", + "zai-org/glm-5:thinking": { + "id": "zai-org/glm-5:thinking", + "name": "GLM 5 Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "structured_output": true, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 4000 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.3, + "output": 2.55 } }, - "o4-mini": { - "id": "o4-mini", - "name": "o4-mini", - "family": "o-mini", + "zai-org/glm-4.6v": { + "id": "zai-org/glm-4.6v", + "name": "GLM 4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "input": 128000, + "output": 24000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.28 + "input": 0.3, + "output": 0.9 } }, - "gpt-5.4-nano": { - "id": "gpt-5.4-nano", - "name": "GPT-5.4 Nano", - "family": "gpt-nano", - "attachment": true, + "zai-org/glm-4.6-original": { + "id": "zai-org/glm-4.6-original", + "name": "GLM 4.6 Original", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-11", + "last_updated": "2025-12-11", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 256000, + "input": 256000, + "output": 65535 + }, + "cost": { + "input": 0.35, + "output": 1.4 + } + }, + "zai-org/glm-5-original:thinking": { + "id": "zai-org/glm-5-original:thinking", + "name": "GLM 5 Original Thinking", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-17", - "last_updated": "2026-03-17", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 200000, + "input": 200000, "output": 128000 }, "cost": { - "input": 0.2, - "output": 1.25, - "cache_read": 0.02 + "input": 1, + "output": 3.2, + "cache_read": 0.2 } }, - "grok-code-fast-1": { - "id": "grok-code-fast-1", - "name": "Grok Code Fast 1", - "family": "grok", + "zai-org/glm-4.7-flash:thinking": { + "id": "zai-org/glm-4.7-flash:thinking", + "name": "GLM 4.7 Flash Thinking", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2025-08-28", - "last_updated": "2025-08-28", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 10000 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 0.2, - "output": 1.5, - "cache_read": 0.02 + "input": 0.07, + "output": 0.4 } }, - "gpt-5.4-pro": { - "id": "gpt-5.4-pro", - "name": "GPT-5.4 Pro", - "family": "gpt-pro", + "zai-org/glm-4.6v-flash-original": { + "id": "zai-org/glm-4.6v-flash-original", + "name": "GLM 4.6V Flash", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", "attachment": true, - "reasoning": true, - "tool_call": true, + "reasoning": false, + "tool_call": false, "structured_output": false, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 24000 }, "cost": { - "input": 30, - "output": 180, - "context_over_200k": { - "input": 60, - "output": 270 - }, - "tiers": [ - { - "input": 60, - "output": 270, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.1, + "output": 0.4 } }, - "o3-mini": { - "id": "o3-mini", - "name": "o3-mini", - "family": "o-mini", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2024-12-20", - "last_updated": "2025-01-29", + "zai-org/glm-4.6v-original": { + "id": "zai-org/glm-4.6v-original", + "name": "GLM 4.6V Original", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 128000, + "input": 128000, + "output": 24000 }, "cost": { - "input": 1.1, - "output": 4.4, - "cache_read": 0.55 + "input": 0.6, + "output": 0.9 } }, - "grok-4": { - "id": "grok-4", - "name": "Grok 4", - "family": "grok", + "zai-org/glm-latest": { + "id": "zai-org/glm-latest", + "name": "GLM Latest", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "structured_output": true, + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 200000, + "input": 200000, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "reasoning": 15, - "cache_read": 0.75 + "input": 0.75, + "output": 2.6, + "cache_read": 0.15 } }, - "gpt-5.4": { - "id": "gpt-5.4", - "name": "GPT-5.4", - "family": "gpt", - "attachment": true, + "zai-org/glm-4.7-flash": { + "id": "zai-org/glm-4.7-flash", + "name": "GLM 4.7 Flash", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1050000, - "input": 922000, + "context": 200000, + "input": 200000, "output": 128000 }, "cost": { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "context_over_200k": { - "input": 5, - "output": 22.5, - "cache_read": 0.5 - }, - "tiers": [ - { - "input": 5, - "output": 22.5, - "cache_read": 0.5, - "tier": { - "type": "context", - "size": 272000 - } - } - ] + "input": 0.07, + "output": 0.4 } }, - "gpt-4.1-nano": { - "id": "gpt-4.1-nano", - "name": "GPT-4.1 nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "zai-org/GLM-4.6-turbo:thinking": { + "id": "zai-org/GLM-4.6-turbo:thinking", + "name": "GLM 4.6 Turbo (Thinking)", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-10-02", + "last_updated": "2025-10-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 200000, + "input": 200000, + "output": 204800 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.03 + "input": 1, + "output": 3 } }, - "grok-3-mini": { - "id": "grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", + "zai-org/GLM-4.5-Air": { + "id": "zai-org/GLM-4.5-Air", + "name": "GLM 4.5 Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "structured_output": true, + "release_date": "2025-04-15", + "last_updated": "2025-04-15", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 98304 }, "cost": { - "input": 0.3, - "output": 0.5, - "reasoning": 0.5, - "cache_read": 0.075 + "input": 0.12, + "output": 0.8 } - }, - "o3": { - "id": "o3", - "name": "o3", - "family": "o", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2024-05", - "release_date": "2025-04-16", - "last_updated": "2025-04-16", + }, + "zai-org/GLM-4.6-turbo": { + "id": "zai-org/GLM-4.6-turbo", + "name": "GLM 4.6 Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-10-02", + "last_updated": "2025-10-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 200000, - "output": 100000 + "input": 200000, + "output": 204800 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 1, + "output": 3 } }, - "gpt-5-pro": { - "id": "gpt-5-pro", - "name": "GPT-5 Pro", - "family": "gpt-pro", - "attachment": true, + "zai-org/glm-5": { + "id": "zai-org/glm-5", + "name": "GLM 5", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "structured_output": true, - "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2025-10-06", - "last_updated": "2025-10-06", + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "output": 272000 + "context": 200000, + "input": 200000, + "output": 128000 }, "cost": { - "input": 15, - "output": 120 + "input": 0.3, + "output": 2.55 } }, - "gpt-4o": { - "id": "gpt-4o", - "name": "GPT-4o", - "family": "gpt", - "attachment": true, + "Infermatic/MN-12B-Inferor-v0.0": { + "id": "Infermatic/MN-12B-Inferor-v0.0", + "name": "Mistral Nemo Inferor 12B", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2024-05-13", - "last_updated": "2024-08-06", + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 2.5, - "output": 10, - "cache_read": 1.25 + "input": 0.25499999999999995, + "output": 0.49299999999999994 } }, - "cohere-command-r-plus-08-2024": { - "id": "cohere-command-r-plus-08-2024", - "name": "Command R+", - "family": "command-r", + "shisa-ai/shisa-v2.1-llama3.3-70b": { + "id": "shisa-ai/shisa-v2.1-llama3.3-70b", + "name": "Shisa V2.1 Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06-01", - "release_date": "2024-08-30", - "last_updated": "2024-08-30", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4000 + "context": 32768, + "input": 32768, + "output": 4096 }, "cost": { - "input": 2.5, - "output": 10 + "input": 0.5, + "output": 0.5 } }, - "gpt-4.1": { - "id": "gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "shisa-ai/shisa-v2-llama3.3-70b": { + "id": "shisa-ai/shisa-v2-llama3.3-70b", + "name": "Shisa V2 Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 128000, + "input": 128000, + "output": 16384 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.5, + "output": 0.5 } }, - "gpt-4.1-mini": { - "id": "gpt-4.1-mini", - "name": "GPT-4.1 mini", - "family": "gpt-mini", - "attachment": true, + "abacusai/Dracarys-72B-Instruct": { + "id": "abacusai/Dracarys-72B-Instruct", + "name": "Llama 3.1 70B Dracarys 2", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-02", + "last_updated": "2025-08-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0.4, - "output": 1.6, - "cache_read": 0.1 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "grok-3": { - "id": "grok-3", - "name": "Grok 3", - "family": "grok", + "deepseek-ai/DeepSeek-V3.1:thinking": { + "id": "deepseek-ai/DeepSeek-V3.1:thinking", + "name": "DeepSeek V3.1 Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek-thinking", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-11", - "release_date": "2025-02-17", - "last_updated": "2025-02-17", + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75 + "input": 0.2, + "output": 0.7 } }, - "grok-4-fast-non-reasoning": { - "id": "grok-4-fast-non-reasoning", - "name": "Grok 4 Fast (Non-Reasoning)", - "family": "grok", - "attachment": true, + "deepseek-ai/DeepSeek-V3.1-Terminus": { + "id": "deepseek-ai/DeepSeek-V3.1-Terminus", + "name": "DeepSeek V3.1 Terminus", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-19", - "last_updated": "2025-09-19", + "structured_output": true, + "release_date": "2025-08-02", + "last_updated": "2025-08-02", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2000000, - "output": 30000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.2, - "output": 0.5, - "cache_read": 0.05 + "input": 0.25, + "output": 0.7 } - } - } - }, - "fastrouter": { - "id": "fastrouter", - "env": ["FASTROUTER_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://go.fastrouter.ai/api/v1", - "name": "FastRouter", - "doc": "https://fastrouter.ai/models", - "models": { - "x-ai/grok-4": { - "id": "x-ai/grok-4", - "name": "Grok 4", - "family": "grok", + }, + "deepseek-ai/deepseek-v3.2-exp-thinking": { + "id": "deepseek-ai/deepseek-v3.2-exp-thinking", + "name": "DeepSeek V3.2 Exp Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 64000 + "context": 163840, + "input": 163840, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.75, - "cache_write": 15 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "deepseek-ai/deepseek-r1-distill-llama-70b": { - "id": "deepseek-ai/deepseek-r1-distill-llama-70b", - "name": "DeepSeek R1 Distill Llama 70B", - "family": "deepseek-thinking", + "deepseek-ai/DeepSeek-R1-0528": { + "id": "deepseek-ai/DeepSeek-R1-0528", + "name": "DeepSeek R1 0528", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-01-23", - "last_updated": "2025-01-23", + "structured_output": false, + "release_date": "2025-05-28", + "last_updated": "2025-05-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 128000, + "input": 128000, + "output": 163840 }, "cost": { - "input": 0.03, - "output": 0.14 + "input": 0.4, + "output": 1.7 } }, - "openai/gpt-5-mini": { - "id": "openai/gpt-5-mini", - "name": "GPT-5 Mini", - "family": "gpt-mini", + "deepseek-ai/DeepSeek-V3.1": { + "id": "deepseek-ai/DeepSeek-V3.1", + "name": "DeepSeek V3.1", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-26", + "last_updated": "2025-07-26", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 2, - "cache_read": 0.025 + "input": 0.2, + "output": 0.7 } }, - "openai/gpt-5-nano": { - "id": "openai/gpt-5-nano", - "name": "GPT-5 Nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "deepseek-ai/deepseek-v3.2-exp": { + "id": "deepseek-ai/deepseek-v3.2-exp", + "name": "DeepSeek V3.2 Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 163840, + "input": 163840, + "output": 65536 }, "cost": { - "input": 0.05, - "output": 0.4, - "cache_read": 0.005 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", + "deepseek-ai/DeepSeek-V3.1-Terminus:thinking": { + "id": "deepseek-ai/DeepSeek-V3.1-Terminus:thinking", + "name": "DeepSeek V3.1 Terminus (Thinking)", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek-thinking", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": true, + "release_date": "2025-09-22", + "last_updated": "2025-09-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 128000, + "input": 128000, "output": 65536 }, "cost": { - "input": 0.05, - "output": 0.2 + "input": 0.25, + "output": 0.7 } }, - "openai/gpt-5": { - "id": "openai/gpt-5", - "name": "GPT-5", - "family": "gpt", - "attachment": true, + "arcee-ai/trinity-large-thinking": { + "id": "arcee-ai/trinity-large-thinking", + "name": "Trinity Large Thinking", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10-01", - "release_date": "2025-08-07", - "last_updated": "2025-08-07", + "structured_output": false, + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "output": 128000 + "context": 262144, + "input": 262144, + "output": 80000 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125 + "input": 0.25, + "output": 0.9 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "arcee-ai/trinity-mini": { + "id": "arcee-ai/trinity-mini", + "name": "Trinity Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "trinity-mini", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, - "output": 32768 + "input": 131072, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0.045000000000000005, + "output": 0.15 } }, - "z-ai/glm-5": { - "id": "z-ai/glm-5", - "name": "GLM-5", - "family": "glm", + "soob3123/GrayLine-Qwen3-8B": { + "id": "soob3123/GrayLine-Qwen3-8B", + "name": "Grayline Qwen3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-25", + "last_updated": "2025-09-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, - "output": 131072 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 0.95, - "output": 3.15 + "input": 0.3, + "output": 0.3 } }, - "qwen/qwen3-coder": { - "id": "qwen/qwen3-coder", - "name": "Qwen3 Coder", - "family": "qwen", + "soob3123/Veiled-Calla-12B": { + "id": "soob3123/Veiled-Calla-12B", + "name": "Veiled Calla 12B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "tool_call": false, + "structured_output": false, + "release_date": "2025-04-13", + "last_updated": "2025-04-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 66536 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { "input": 0.3, - "output": 1.2 + "output": 0.3 } }, - "google/gemini-2.5-pro": { - "id": "google/gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "soob3123/amoral-gemma3-27B-v2": { + "id": "soob3123/amoral-gemma3-27B-v2", + "name": "Amoral Gemma3 27B v2", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-05-23", + "last_updated": "2025-05-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 32768, + "input": 32768, + "output": 8192 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 0.3, + "output": 0.3 } }, - "google/gemini-2.5-flash": { - "id": "google/gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "meganova-ai/manta-mini-1.0": { + "id": "meganova-ai/manta-mini-1.0", + "name": "Manta Mini 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-20", + "last_updated": "2025-12-20", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.0375 + "input": 0.02, + "output": 0.16 } }, - "moonshotai/kimi-k2": { - "id": "moonshotai/kimi-k2", - "name": "Kimi K2", - "family": "kimi", + "meganova-ai/manta-flash-1.0": { + "id": "meganova-ai/manta-flash-1.0", + "name": "Manta Flash 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-11", - "last_updated": "2025-07-11", + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-20", + "last_updated": "2025-12-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 0.02, + "output": 0.16 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", - "attachment": true, + "meganova-ai/manta-pro-1.0": { + "id": "meganova-ai/manta-pro-1.0", + "name": "Manta Pro 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-20", + "last_updated": "2025-12-20", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1047576, + "context": 32768, + "input": 32768, "output": 32768 }, "cost": { - "input": 2, - "output": 8, - "cache_read": 0.5 + "input": 0.060000000000000005, + "output": 0.5 } }, - "anthropic/claude-opus-4.1": { - "id": "anthropic/claude-opus-4.1", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "qwen/Qwen3.6-35B-A3B:thinking": { + "id": "qwen/Qwen3.6-35B-A3B:thinking", + "name": "Qwen3.6 35B A3B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-19", + "last_updated": "2026-04-19", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 32000 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.112, + "output": 0.8 } }, - "anthropic/claude-sonnet-4": { - "id": "anthropic/claude-sonnet-4", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "qwen/qwen3-coder-plus": { + "id": "qwen/qwen3-coder-plus", + "name": "Qwen3 Coder Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-17", + "last_updated": "2025-09-17", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 64000 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1, + "output": 5 } - } - } - }, - "stackit": { - "id": "stackit", - "env": ["STACKIT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.openai-compat.model-serving.eu01.onstackit.cloud/v1", - "name": "STACKIT", - "doc": "https://docs.stackit.cloud/products/data-and-ai/ai-model-serving/basics/available-shared-models", - "models": { - "Qwen/Qwen3-VL-Embedding-8B": { - "id": "Qwen/Qwen3-VL-Embedding-8B", - "name": "Qwen3-VL Embedding 8B", - "family": "qwen", + }, + "qwen/Qwen3.6-35B-A3B": { + "id": "qwen/Qwen3.6-35B-A3B", + "name": "Qwen3.6 35B A3B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": false, - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 16384 }, "cost": { - "input": 0.09, - "output": 0.09 + "input": 0.112, + "output": 0.8 } }, - "Qwen/Qwen3-VL-235B-A22B-Instruct-FP8": { - "id": "Qwen/Qwen3-VL-235B-A22B-Instruct-FP8", - "name": "Qwen3-VL 235B", - "family": "qwen", + "qwen/qwen3-32b": { + "id": "qwen/qwen3-32b", + "name": "Qwen 3 32b", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 218000, - "output": 8192 + "context": 41000, + "input": 41000, + "output": 32768 }, "cost": { - "input": 1.64, - "output": 1.91 + "input": 0.1, + "output": 0.3 } }, - "neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8": { - "id": "neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8", - "name": "Llama 3.1 8B", - "family": "llama", - "attachment": false, + "qwen/Qwen3-VL-235B-A22B-Instruct": { + "id": "qwen/Qwen3-VL-235B-A22B-Instruct", + "name": "Qwen3 VL 235B A22B Instruct", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "input": 128000, + "output": 262144 }, "cost": { - "input": 0.16, - "output": 0.27 + "input": 0.3, + "output": 1.2 } }, - "neuralmagic/Mistral-Nemo-Instruct-2407-FP8": { - "id": "neuralmagic/Mistral-Nemo-Instruct-2407-FP8", - "name": "Mistral Nemo", - "family": "mistral", + "qwen/qwen3-max": { + "id": "qwen/qwen3-max", + "name": "Qwen3 Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "attachment": false, "reasoning": false, - "tool_call": true, + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2024-07-01", - "last_updated": "2024-07-01", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 32768 }, "cost": { - "input": 0.49, - "output": 0.71 + "input": 1.08018, + "output": 5.4009 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT-OSS 120B", - "family": "gpt", - "attachment": false, + "qwen/qwen3.5-397b-a17b-thinking": { + "id": "qwen/qwen3.5-397b-a17b-thinking", + "name": "Qwen3.5 397B A17B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, "reasoning": true, - "tool_call": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, "structured_output": false, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131000, - "output": 8192 + "context": 258048, + "input": 258048, + "output": 65536 }, "cost": { - "input": 0.49, - "output": 0.71 + "input": 0.6, + "output": 3.6 } }, - "cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic": { - "id": "cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic", - "name": "Llama 3.3 70B", - "family": "llama", + "qwen/Qwen3-Next-80B-A3B-Instruct": { + "id": "qwen/Qwen3-Next-80B-A3B-Instruct", + "name": "Qwen3 Next 80B A3B (Instruct)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, "tool_call": true, - "structured_output": false, - "temperature": true, - "release_date": "2024-12-05", - "last_updated": "2024-12-05", + "structured_output": true, + "release_date": "2025-09-11", + "last_updated": "2025-09-11", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 0.49, - "output": 0.71 + "input": 0.15, + "output": 0.65 } }, - "google/gemma-3-27b-it": { - "id": "google/gemma-3-27b-it", - "name": "Gemma 3 27B", - "family": "gemma", - "attachment": true, + "qwen/Qwen3-235B-A22B-Instruct-2507-TEE": { + "id": "qwen/Qwen3-235B-A22B-Instruct-2507-TEE", + "name": "Qwen 3 235b A22B 2507 (TEE)", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, "reasoning": false, - "tool_call": false, - "structured_output": false, - "temperature": true, - "release_date": "2025-05-17", - "last_updated": "2025-05-17", + "tool_call": true, + "structured_output": true, + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 37000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 0.49, - "output": 0.71 + "input": 0.13, + "output": 0.5 } }, - "intfloat/e5-mistral-7b-instruct": { - "id": "intfloat/e5-mistral-7b-instruct", - "name": "E5 Mistral 7B", - "family": "mistral", + "qwen/qwq-32b-preview": { + "id": "qwen/qwq-32b-preview", + "name": "Qwen QwQ 32B Preview", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": false, - "release_date": "2023-12-11", - "last_updated": "2023-12-11", + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 4096 + "context": 32768, + "input": 32768, + "output": 32768 }, "cost": { - "input": 0.02, - "output": 0.02 + "input": 0.2, + "output": 0.2 } - } - } - }, - "tencent-coding-plan": { - "id": "tencent-coding-plan", - "env": ["TENCENT_CODING_PLAN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.lkeap.cloud.tencent.com/coding/v3", - "name": "Tencent Coding Plan (China)", - "doc": "https://cloud.tencent.com/document/product/1772/128947", - "models": { - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi-K2.5", - "family": "kimi", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + }, + "qwen/qwen3-next-80b-a3b-thinking": { + "id": "qwen/qwen3-next-80b-a3b-thinking", + "name": "Qwen3 Next 80B A3B (Thinking)", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, + "context": 256000, + "input": 256000, "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.15, + "output": 0.65 } }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", + "qwen/qwen3-coder": { + "id": "qwen/qwen3-coder", + "name": "Qwen 3 Coder 480B", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", + "structured_output": true, + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 202752, - "output": 16384 + "context": 262000, + "input": 262000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.13, + "output": 0.5 } }, - "hunyuan-turbos": { - "id": "hunyuan-turbos", - "name": "Hunyuan-TurboS", - "family": "hunyuan", + "qwen/Qwen3-235B-A22B-Thinking-2507": { + "id": "qwen/Qwen3-235B-A22B-Thinking-2507", + "name": "Qwen 3 235b A22B 2507 Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-08", - "last_updated": "2026-03-08", + "tool_call": false, + "structured_output": false, + "release_date": "2025-09-11", + "last_updated": "2025-09-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.3, + "output": 0.5 } }, - "hunyuan-t1": { - "id": "hunyuan-t1", - "name": "Hunyuan-T1", - "family": "hunyuan", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-03-08", - "last_updated": "2026-03-08", + "qwen/qwen3.5-plus": { + "id": "qwen/qwen3.5-plus", + "name": "Qwen3.5 Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 983616, + "input": 983616, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.4, + "output": 2.4, + "cache_read": 0.04 } }, - "hunyuan-2.0-instruct": { - "id": "hunyuan-2.0-instruct", - "name": "Tencent HY 2.0 Instruct", - "family": "hunyuan", - "attachment": false, - "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-08", - "last_updated": "2026-03-08", + "qwen/qwen3.5-plus-thinking": { + "id": "qwen/qwen3.5-plus-thinking", + "name": "Qwen3.5 Plus Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 983616, + "input": 983616, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.4, + "output": 2.4, + "cache_read": 0.04 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, + "qwen/qwen3-235b-a22b": { + "id": "qwen/qwen3-235b-a22b", + "name": "Qwen 3 235b A22B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "structured_output": true, + "release_date": "2025-04-29", + "last_updated": "2025-04-29", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 204800, + "context": 41000, + "input": 41000, "output": 32768 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.3, + "output": 0.5 } }, - "tc-code-latest": { - "id": "tc-code-latest", - "name": "Auto", - "family": "auto", + "qwen/qwen-2.5-72b-instruct": { + "id": "qwen/qwen-2.5-72b-instruct", + "name": "Qwen2.5 72B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "release_date": "2026-03-08", - "last_updated": "2026-03-08", + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-03", + "last_updated": "2025-07-03", "modalities": { "input": ["text"], "output": ["text"] @@ -110070,3368 +164266,3708 @@ "open_weights": false, "limit": { "context": 131072, - "output": 16384 + "input": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.357, + "output": 0.408 } }, - "hunyuan-2.0-thinking": { - "id": "hunyuan-2.0-thinking", - "name": "Tencent HY 2.0 Think", - "family": "hunyuan", + "qwen/qwen3-coder-next": { + "id": "qwen/qwen3-coder-next", + "name": "Qwen3 Coder Next", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-03-08", - "last_updated": "2026-03-08", + "structured_output": true, + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "input": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.15, + "output": 1.5 } - } - } - }, - "privatemode-ai": { - "id": "privatemode-ai", - "env": ["PRIVATEMODE_API_KEY", "PRIVATEMODE_ENDPOINT"], - "npm": "@ai-sdk/openai-compatible", - "api": "http://localhost:8080/v1", - "name": "Privatemode AI", - "doc": "https://docs.privatemode.ai/api/overview", - "models": { - "gemma-3-27b": { - "id": "gemma-3-27b", - "name": "Gemma 3 27B", - "family": "gemma", + }, + "qwen/qwen3.5-9b": { + "id": "qwen/qwen3.5-9b", + "name": "Qwen3.5 9B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 81920 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-10", + "last_updated": "2026-03-10", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 256000, + "input": 256000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.05, + "output": 0.15 } }, - "whisper-large-v3": { - "id": "whisper-large-v3", - "name": "Whisper large-v3", - "family": "whisper", - "attachment": true, + "qwen/qwen3.5-397b-a17b": { + "id": "qwen/qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "knowledge": "2023-09", - "release_date": "2023-09-01", - "last_updated": "2023-09-01", + "release_date": "2026-02-16", + "last_updated": "2026-02-16", "modalities": { - "input": ["audio"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 0, - "output": 4096 + "context": 258048, + "input": 258048, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 3.6 } }, - "qwen3-embedding-4b": { - "id": "qwen3-embedding-4b", - "name": "Qwen3-Embedding 4B", - "family": "qwen", + "qwen/qwen3-coder-flash": { + "id": "qwen/qwen3-coder-flash", + "name": "Qwen3 Coder Flash", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, "reasoning": false, "tool_call": false, "structured_output": false, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-06-06", - "last_updated": "2025-06-06", + "release_date": "2025-09-17", + "last_updated": "2025-09-17", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 2560 + "context": 128000, + "input": 128000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.5 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "gpt-oss-120b", - "family": "gpt-oss", + "qwen/Qwen2.5-Coder-32B-Instruct": { + "id": "qwen/Qwen2.5-Coder-32B-Instruct", + "name": "Qwen 2.5 Coder 32b", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-04", - "last_updated": "2025-08-14", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-03", + "last_updated": "2025-07-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 128000 + "context": 32000, + "input": 32000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2006, + "output": 0.2006 } }, - "qwen3-coder-30b-a3b": { - "id": "qwen3-coder-30b-a3b", - "name": "Qwen3-Coder 30B-A3B", - "family": "qwen", + "qwen/Qwen3-8B": { + "id": "qwen/Qwen3-8B", + "name": "Qwen 3 8B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04", - "last_updated": "2025-04", + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 41000, + "input": 41000, "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.47, + "output": 0.47 } - } - } - }, - "google": { - "id": "google", - "env": ["GOOGLE_GENERATIVE_AI_API_KEY", "GEMINI_API_KEY"], - "npm": "@ai-sdk/google", - "name": "Google", - "doc": "https://ai.google.dev/gemini-api/docs/models", - "models": { - "gemini-flash-lite-latest": { - "id": "gemini-flash-lite-latest", - "name": "Gemini Flash-Lite Latest", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + }, + "qwen/qwen3-14b": { + "id": "qwen/qwen3-14b", + "name": "Qwen 3 14b", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-01-01", + "last_updated": "2024-01-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 41000, + "input": 41000, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.08, + "output": 0.24 } }, - "gemini-2.5-pro-preview-05-06": { - "id": "gemini-2.5-pro-preview-05-06", - "name": "Gemini 2.5 Pro Preview 05-06", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, + "qwen/Qwen3-235B-A22B-Instruct-2507": { + "id": "qwen/Qwen3-235B-A22B-Instruct-2507", + "name": "Qwen 3 235b A22B 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "attachment": false, + "reasoning": false, "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-05-06", - "last_updated": "2025-05-06", + "release_date": "2025-07-25", + "last_updated": "2025-07-25", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 256000, + "input": 256000, + "output": 262144 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 0.13, + "output": 0.5 } }, - "gemini-live-2.5-flash-preview-native-audio": { - "id": "gemini-live-2.5-flash-preview-native-audio", - "name": "Gemini Live 2.5 Flash Preview Native Audio", - "family": "gemini-flash", + "qwen/qwen3-30b-a3b": { + "id": "qwen/qwen3-30b-a3b", + "name": "Qwen3 30B A3B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-09-18", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-27", + "last_updated": "2025-02-27", "modalities": { - "input": ["text", "audio", "video"], - "output": ["text", "audio"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 65536 + "context": 41000, + "input": 41000, + "output": 32768 }, "cost": { - "input": 0.5, - "output": 2, - "input_audio": 3, - "output_audio": 12 + "input": 0.1, + "output": 0.3 } }, - "gemini-3.1-pro-preview-customtools": { - "id": "gemini-3.1-pro-preview-customtools", - "name": "Gemini 3.1 Pro Preview Custom Tools", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "amazon/nova-lite-v1": { + "id": "amazon/nova-lite-v1", + "name": "Amazon Nova Lite 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-lite", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 300000, + "input": 300000, + "output": 5120 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.0595, + "output": 0.238 } }, - "gemini-2.5-flash-lite-preview-09-2025": { - "id": "gemini-2.5-flash-lite-preview-09-2025", - "name": "Gemini 2.5 Flash Lite Preview 09-25", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "amazon/nova-pro-v1": { + "id": "amazon/nova-pro-v1", + "name": "Amazon Nova Pro 1.0", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "nova-pro", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 300000, + "input": 300000, + "output": 32000 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.7989999999999999, + "output": 3.1959999999999997 } }, - "gemini-1.5-flash": { - "id": "gemini-1.5-flash", - "name": "Gemini 1.5 Flash", - "family": "gemini-flash", - "attachment": true, + "amazon/nova-micro-v1": { + "id": "amazon/nova-micro-v1", + "name": "Amazon Nova Micro 1.0", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova-micro", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-05-14", - "last_updated": "2024-05-14", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 5120 }, "cost": { - "input": 0.075, - "output": 0.3, - "cache_read": 0.01875 + "input": 0.0357, + "output": 0.1394 } }, - "gemini-1.5-pro": { - "id": "gemini-1.5-pro", - "name": "Gemini 1.5 Pro", - "family": "gemini-pro", - "attachment": true, + "amazon/nova-2-lite-v1": { + "id": "amazon/nova-2-lite-v1", + "name": "Amazon Nova 2 Lite", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "nova", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-02-15", - "last_updated": "2024-02-15", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-03", + "last_updated": "2024-12-03", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1000000, - "output": 8192 + "input": 1000000, + "output": 65535 }, "cost": { - "input": 1.25, - "output": 5, - "cache_read": 0.3125 + "input": 0.5099999999999999, + "output": 4.25 } }, - "gemma-3n-e4b-it": { - "id": "gemma-3n-e4b-it", - "name": "Gemma 3n 4B", - "family": "gemma", - "attachment": true, + "alibaba/qwen3.6-flash": { + "id": "alibaba/qwen3.6-flash", + "name": "Qwen3.6 Flash", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen3.6", + "attachment": false, "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "structured_output": false, + "release_date": "2026-04-17", + "last_updated": "2026-04-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2000 + "context": 991800, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.19, + "output": 1.16 } }, - "gemini-3.1-flash-lite-preview": { - "id": "gemini-3.1-flash-lite-preview", - "name": "Gemini 3.1 Flash Lite Preview", - "family": "gemini-flash-lite", + "alibaba/qwen3.6-27b": { + "id": "alibaba/qwen3.6-27b", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-03-03", - "last_updated": "2026-03-03", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 260096, + "input": 260096, "output": 65536 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 0.203, + "output": 2.24 } }, - "gemini-3.1-pro-preview": { - "id": "gemini-3.1-pro-preview", - "name": "Gemini 3.1 Pro Preview", - "family": "gemini-pro", + "alibaba/qwen3.6-27b:thinking": { + "id": "alibaba/qwen3.6-27b:thinking", + "name": "Qwen3.6 27B Thinking", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-19", - "last_updated": "2026-02-19", + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 131072 + } + ], + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 260096, + "input": 260096, "output": 65536 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.203, + "output": 2.24 } }, - "gemini-2.0-flash": { - "id": "gemini-2.0-flash", - "name": "Gemini 2.0 Flash", - "family": "gemini-flash", - "attachment": true, + "aion-labs/aion-rp-llama-3.1-8b": { + "id": "aion-labs/aion-rp-llama-3.1-8b", + "name": "Llama 3.1 8b (uncensored)", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 32768, + "input": 32768, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025 + "input": 0.2006, + "output": 0.2006 } }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "aion-labs/aion-2.5": { + "id": "aion-labs/aion-2.5", + "name": "AionLabs: Aion-2.5", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-20", + "last_updated": "2026-03-20", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 131072, + "input": 131072, + "output": 32768 }, "cost": { - "input": 0.5, + "input": 1, "output": 3, - "cache_read": 0.05, - "input_audio": 1 + "cache_read": 0.35 } }, - "gemini-2.5-flash-preview-tts": { - "id": "gemini-2.5-flash-preview-tts", - "name": "Gemini 2.5 Flash Preview TTS", - "family": "gemini-flash", + "aion-labs/aion-1.0-mini": { + "id": "aion-labs/aion-1.0-mini", + "name": "Aion 1.0 mini (DeepSeek)", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-05-01", - "last_updated": "2025-05-01", + "structured_output": false, + "release_date": "2025-02-20", + "last_updated": "2025-02-20", "modalities": { "input": ["text"], - "output": ["audio"] - }, - "open_weights": false, - "limit": { - "context": 8000, - "output": 16000 - }, - "cost": { - "input": 0.5, - "output": 10 - } - }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-11-18", - "last_updated": "2025-11-18", - "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 131072, + "input": 131072, + "output": 8192 }, "cost": { - "input": 2, - "output": 12, - "cache_read": 0.2, - "tiers": [ - { - "input": 4, - "output": 18, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 4, - "output": 18, - "cache_read": 0.4 - } + "input": 0.7989999999999999, + "output": 1.394 } }, - "gemini-2.5-flash-preview-05-20": { - "id": "gemini-2.5-flash-preview-05-20", - "name": "Gemini 2.5 Flash Preview 05-20", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "aion-labs/aion-2.0": { + "id": "aion-labs/aion-2.0", + "name": "AionLabs: Aion-2.0", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 131072, + "input": 131072, + "output": 32768 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.0375 + "input": 0.8, + "output": 1.6 } }, - "gemini-embedding-001": { - "id": "gemini-embedding-001", - "name": "Gemini Embedding 001", - "family": "gemini", + "aion-labs/aion-1.0": { + "id": "aion-labs/aion-1.0", + "name": "Aion 1.0", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2025-05", - "release_date": "2025-05-20", - "last_updated": "2025-05-20", + "structured_output": false, + "release_date": "2025-02-01", + "last_updated": "2025-02-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 2048, - "output": 3072 + "context": 65536, + "input": 65536, + "output": 8192 }, "cost": { - "input": 0.15, - "output": 0 + "input": 3.995, + "output": 7.99 } }, - "gemini-2.5-pro": { - "id": "gemini-2.5-pro", - "name": "Gemini 2.5 Pro", - "family": "gemini-pro", - "attachment": true, + "pamanseau/OpenReasoning-Nemotron-32B": { + "id": "pamanseau/OpenReasoning-Nemotron-32B", + "name": "OpenReasoning Nemotron 32B", + "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents", + "family": "nemotron", + "attachment": false, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 32768, + "input": 32768, "output": 65536 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.125, - "tiers": [ - { - "input": 2.5, - "output": 15, - "cache_read": 0.25, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 2.5, - "output": 15, - "cache_read": 0.25 - } + "input": 0.1, + "output": 0.4 } }, - "gemini-flash-latest": { - "id": "gemini-flash-latest", - "name": "Gemini Flash Latest", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "LatitudeGames/Wayfarer-Large-70B-Llama-3.3": { + "id": "LatitudeGames/Wayfarer-Large-70B-Llama-3.3", + "name": "Llama 3.3 70B Wayfarer", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-20", + "last_updated": "2025-02-20", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "input_audio": 1 + "input": 0.700000007, + "output": 0.700000007 } }, - "gemma-4-31b-it": { - "id": "gemma-4-31b-it", - "name": "Gemma 4 31B", - "family": "gemma", + "baseten/Kimi-K2-Instruct-FP4": { + "id": "baseten/Kimi-K2-Instruct-FP4", + "name": "Kimi K2 0711 Instruct FP4", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-11", + "last_updated": "2025-07-11", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 128000, + "input": 128000, + "output": 131072 + }, + "cost": { + "input": 0.1, + "output": 2 } }, - "gemini-2.5-pro-preview-06-05": { - "id": "gemini-2.5-pro-preview-06-05", - "name": "Gemini 2.5 Pro Preview 06-05", - "family": "gemini-pro", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-05", - "last_updated": "2025-06-05", + "inflection/inflection-3-pi": { + "id": "inflection/inflection-3-pi", + "name": "Inflection 3 Pi", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-10-11", + "last_updated": "2024-10-11", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 8000, + "input": 8000, + "output": 4096 }, "cost": { - "input": 1.25, - "output": 10, - "cache_read": 0.31 + "input": 2.499, + "output": 9.996 } }, - "gemini-2.5-flash-image": { - "id": "gemini-2.5-flash-image", - "name": "Gemini 2.5 Flash Image", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, + "inflection/inflection-3-productivity": { + "id": "inflection/inflection-3-productivity", + "name": "Inflection 3 Productivity", + "description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks", + "family": "gpt", + "attachment": false, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "structured_output": false, + "release_date": "2024-10-11", + "last_updated": "2024-10-11", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 8000, + "input": 8000, + "output": 4096 }, "cost": { - "input": 0.3, - "output": 30, - "cache_read": 0.075 + "input": 2.499, + "output": 9.996 } }, - "gemini-2.5-flash-lite-preview-06-17": { - "id": "gemini-2.5-flash-lite-preview-06-17", - "name": "Gemini 2.5 Flash Lite Preview 06-17", - "family": "gemini-flash-lite", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0": { + "id": "ReadyArt/MS3.2-The-Omega-Directive-24B-Unslop-v2.0", + "name": "Omega Directive 24B Unslop v2.0", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 32768 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.025, - "input_audio": 0.3 + "input": 0.5, + "output": 0.5 } }, - "gemma-3-12b-it": { - "id": "gemma-3-12b-it", - "name": "Gemma 3 12B", - "family": "gemma", - "attachment": true, + "MiniMaxAI/MiniMax-M1-80k": { + "id": "MiniMaxAI/MiniMax-M1-80k", + "name": "MiniMax M1 80K", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": false, "tool_call": false, - "structured_output": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "structured_output": false, + "release_date": "2025-06-16", + "last_updated": "2025-06-16", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 1000000, + "input": 1000000, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6052, + "output": 2.4225000000000003 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-03-20", - "last_updated": "2025-06-05", + "upstage/solar-pro-3": { + "id": "upstage/solar-pro-3", + "name": "Solar Pro 3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 65536 + "context": 128000, + "input": 128000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.03, - "input_audio": 1 + "input": 0.15, + "output": 0.6, + "cache_read": 0.015 } }, - "gemma-3n-e2b-it": { - "id": "gemma-3n-e2b-it", - "name": "Gemma 3n 2B", - "family": "gemma", - "attachment": true, - "reasoning": false, + "allenai/olmo-3-32b-think": { + "id": "allenai/olmo-3-32b-think", + "name": "Olmo 3 32B Think", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", + "family": "allenai", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-09", - "last_updated": "2025-07-09", + "structured_output": false, + "release_date": "2025-11-01", + "last_updated": "2025-11-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2000 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 0.44999999999999996 } }, - "gemini-3.1-flash-image-preview": { - "id": "gemini-3.1-flash-image-preview", - "name": "Gemini 3.1 Flash Image (Preview)", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, + "essentialai/rnj-1-instruct": { + "id": "essentialai/rnj-1-instruct", + "name": "RNJ-1 Instruct 8B", + "description": "General-purpose chat model for instruction following, writing, and analysis", + "family": "rnj", + "attachment": false, + "reasoning": false, "tool_call": false, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-02-26", - "last_updated": "2026-02-26", + "structured_output": false, + "release_date": "2025-12-13", + "last_updated": "2025-12-13", "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 131072, - "output": 32768 + "context": 128000, + "input": 128000, + "output": 8192 }, "cost": { - "input": 0.5, - "output": 60 + "input": 0.15, + "output": 0.15 } }, - "gemini-3.1-flash-lite": { - "id": "gemini-3.1-flash-lite", - "name": "Gemini 3.1 Flash Lite", - "family": "gemini-flash-lite", - "attachment": true, + "deepseek/deepseek-v4-flash": { + "id": "deepseek/deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-05-07", - "last_updated": "2026-05-07", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 65536 + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0.25, - "output": 1.5, - "cache_read": 0.025, - "input_audio": 0.5 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "gemma-3-4b-it": { - "id": "gemma-3-4b-it", - "name": "Gemma 3 4B", - "family": "gemma", - "attachment": true, - "reasoning": false, - "tool_call": false, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-13", - "last_updated": "2025-03-13", + "deepseek/deepseek-v4-flash:thinking": { + "id": "deepseek/deepseek-v4-flash:thinking", + "name": "DeepSeek V4 Flash (Thinking)", + "description": "Fast DeepSeek model for efficient chat, coding help, and agent loops", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32768, - "output": 8192 + "context": 1048576, + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.14, + "output": 0.28, + "cache_read": 0.028 } }, - "gemini-2.5-flash-preview-04-17": { - "id": "gemini-2.5-flash-preview-04-17", - "name": "Gemini 2.5 Flash Preview 04-17", - "family": "gemini-flash", - "attachment": true, + "deepseek/deepseek-v4-pro-cheaper:thinking": { + "id": "deepseek/deepseek-v4-pro-cheaper:thinking", + "name": "DeepSeek V4 Pro Cheaper (Thinking)", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-04-17", - "last_updated": "2025-04-17", + "structured_output": true, + "release_date": "2026-04-25", + "last_updated": "2026-04-25", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 65536 + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0.15, - "output": 0.6, - "cache_read": 0.0375 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "gemini-2.5-pro-preview-tts": { - "id": "gemini-2.5-pro-preview-tts", - "name": "Gemini 2.5 Pro Preview TTS", - "family": "gemini-flash", + "deepseek/deepseek-prover-v2-671b": { + "id": "deepseek/deepseek-prover-v2-671b", + "name": "DeepSeek Prover v2 671B", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2025-05-01", - "last_updated": "2025-05-01", + "structured_output": false, + "release_date": "2025-04-30", + "last_updated": "2025-04-30", "modalities": { "input": ["text"], - "output": ["audio"] + "output": ["text"] }, "open_weights": false, "limit": { - "context": 8000, - "output": 16000 + "context": 160000, + "input": 160000, + "output": 16384 }, "cost": { "input": 1, - "output": 20 + "output": 2.5 } }, - "gemini-2.5-flash-preview-09-2025": { - "id": "gemini-2.5-flash-preview-09-2025", - "name": "Gemini 2.5 Flash Preview 09-25", - "family": "gemini-flash", - "attachment": true, + "deepseek/deepseek-latest": { + "id": "deepseek/deepseek-latest", + "name": "DeepSeek Latest", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-25", - "last_updated": "2025-09-25", + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 1048576, - "output": 65536 + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0.3, - "output": 2.5, - "cache_read": 0.075, - "input_audio": 1 + "input": 1.1, + "output": 2.2, + "cache_read": 0.11 } }, - "gemma-3-27b-it": { - "id": "gemma-3-27b-it", - "name": "Gemma 3 27B", - "family": "gemma", - "attachment": true, - "reasoning": false, + "deepseek/deepseek-v4-pro": { + "id": "deepseek/deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-03-12", - "last_updated": "2025-03-12", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 1048576, + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.1, + "output": 2.2, + "cache_read": 0.11 } }, - "gemma-4-26b-a4b-it": { - "id": "gemma-4-26b-a4b-it", - "name": "Gemma 4 26B", - "family": "gemma", + "deepseek/deepseek-v4-pro:thinking": { + "id": "deepseek/deepseek-v4-pro:thinking", + "name": "DeepSeek V4 Pro (Thinking)", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, - "temperature": true, - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 256000, - "output": 8192 + "context": 1048576, + "input": 1048576, + "output": 384000 + }, + "cost": { + "input": 1.1, + "output": 2.2, + "cache_read": 0.11 } }, - "gemini-2.5-flash-lite": { - "id": "gemini-2.5-flash-lite", - "name": "Gemini 2.5 Flash Lite", - "family": "gemini-flash-lite", + "deepseek/deepseek-v3.2-speciale": { + "id": "deepseek/deepseek-v3.2-speciale", + "name": "DeepSeek V3.2 Speciale", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": true, "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-06-17", - "last_updated": "2025-06-17", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-02", + "last_updated": "2025-12-02", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, + "context": 163000, + "input": 163000, "output": 65536 }, "cost": { - "input": 0.1, - "output": 0.4, - "cache_read": 0.01, - "input_audio": 0.3 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "gemini-2.5-flash-image-preview": { - "id": "gemini-2.5-flash-image-preview", - "name": "Gemini 2.5 Flash Image (Preview)", - "family": "gemini-flash", - "attachment": true, + "deepseek/deepseek-v4-pro-cheaper": { + "id": "deepseek/deepseek-v4-pro-cheaper", + "name": "DeepSeek V4 Pro Cheaper", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", + "attachment": false, "reasoning": true, - "tool_call": false, - "temperature": true, - "knowledge": "2025-06", - "release_date": "2025-08-26", - "last_updated": "2025-08-26", + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "xhigh"] + } + ], + "tool_call": true, + "structured_output": true, + "release_date": "2026-04-25", + "last_updated": "2026-04-25", "modalities": { - "input": ["text", "image"], - "output": ["text", "image"] + "input": ["text"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 32768 + "context": 1048576, + "input": 1048576, + "output": 384000 }, "cost": { - "input": 0.3, - "output": 30, - "cache_read": 0.075 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "gemini-1.5-flash-8b": { - "id": "gemini-1.5-flash-8b", - "name": "Gemini 1.5 Flash-8B", - "family": "gemini-flash", + "deepseek/deepseek-v3.2:thinking": { + "id": "deepseek/deepseek-v3.2:thinking", + "name": "DeepSeek V3.2 Thinking", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-04", - "release_date": "2024-10-03", - "last_updated": "2024-10-03", + "structured_output": true, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 8192 + "context": 163000, + "input": 163000, + "output": 65536 }, "cost": { - "input": 0.0375, - "output": 0.15, - "cache_read": 0.01 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "gemini-live-2.5-flash": { - "id": "gemini-live-2.5-flash", - "name": "Gemini Live 2.5 Flash", - "family": "gemini-flash", + "deepseek/deepseek-v3.2": { + "id": "deepseek/deepseek-v3.2", + "name": "DeepSeek V3.2", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-09-01", - "last_updated": "2025-09-01", + "structured_output": true, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text", "audio"] + "input": ["text", "pdf"], + "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8000 + "context": 163000, + "input": 163000, + "output": 65536 }, "cost": { - "input": 0.5, - "output": 2, - "input_audio": 3, - "output_audio": 12 + "input": 0.27999999999999997, + "output": 0.42000000000000004 } }, - "gemini-2.0-flash-lite": { - "id": "gemini-2.0-flash-lite", - "name": "Gemini 2.0 Flash Lite", - "family": "gemini-flash-lite", + "minimax/minimax-m3:thinking": { + "id": "minimax/minimax-m3:thinking", + "name": "MiniMax M3 Thinking", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1048576, - "output": 8192 + "context": 512000, + "input": 512000, + "output": 80000 }, "cost": { - "input": 0.075, - "output": 0.3 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } - } - } - }, - "drun": { - "id": "drun", - "env": ["DRUN_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://chat.d.run/v1", - "name": "D.Run (China)", - "doc": "https://www.d.run", - "models": { - "public/deepseek-r1": { - "id": "public/deepseek-r1", - "name": "DeepSeek R1", - "family": "deepseek-thinking", + }, + "minimax/minimax-m2.5": { + "id": "minimax/minimax-m2.5", + "name": "MiniMax M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "structured_output": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 32000 + "context": 204800, + "input": 204800, + "output": 131072 }, "cost": { - "input": 0.55, - "output": 2.2 + "input": 0.3, + "output": 1.2 } }, - "public/minimax-m25": { - "id": "public/minimax-m25", - "name": "MiniMax M2.5", + "minimax/minimax-m2.1": { + "id": "minimax/minimax-m2.1", + "name": "MiniMax M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_details" - }, - "temperature": true, - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "structured_output": true, + "release_date": "2025-12-19", + "last_updated": "2025-12-19", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 204800, + "context": 200000, + "input": 200000, "output": 131072 }, "cost": { - "input": 0.29, - "output": 1.16 + "input": 0.33, + "output": 1.32 } }, - "public/deepseek-v3": { - "id": "public/deepseek-v3", - "name": "DeepSeek V3", - "family": "deepseek", + "minimax/minimax-m2.7-turbo": { + "id": "minimax/minimax-m2.7-turbo", + "name": "MiniMax M2.7 Turbo", + "description": "Efficient MiniMax model for quick assistance, coding, and routine automation", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-12-26", - "last_updated": "2024-12-26", + "structured_output": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 8192 + "context": 204800, + "input": 204800, + "output": 131072 }, "cost": { - "input": 0.28, - "output": 1.1 + "input": 0.6, + "output": 2.4 } - } - } - }, - "moonshotai": { - "id": "moonshotai", - "env": ["MOONSHOT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.moonshot.ai/v1", - "name": "Moonshot AI", - "doc": "https://platform.moonshot.ai/docs/api/chat", - "models": { - "kimi-k2-0905-preview": { - "id": "kimi-k2-0905-preview", - "name": "Kimi K2 0905", - "family": "kimi", - "attachment": false, + }, + "minimax/minimax-01": { + "id": "minimax/minimax-01", + "name": "MiniMax 01", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", + "attachment": true, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "tool_call": false, + "structured_output": false, + "release_date": "2025-01-15", + "last_updated": "2025-01-15", "modalities": { - "input": ["text"], + "input": ["text", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000192, + "input": 1000192, + "output": 16384 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.1394, + "output": 1.1219999999999999 } }, - "kimi-k2.5": { - "id": "kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi-k2.5", - "attachment": false, - "reasoning": true, + "minimax/minimax-m3": { + "id": "minimax/minimax-m3", + "name": "MiniMax M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "attachment": true, + "reasoning": false, "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": false, - "knowledge": "2025-01", - "release_date": "2026-01", - "last_updated": "2026-01", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 512000, + "input": 512000, + "output": 80000 }, "cost": { - "input": 0.6, - "output": 3, - "cache_read": 0.1 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "kimi-k2-thinking-turbo": { - "id": "kimi-k2-thinking-turbo", - "name": "Kimi K2 Thinking Turbo", - "family": "kimi-thinking", + "minimax/minimax-m2-her": { + "id": "minimax/minimax-m2-her", + "name": "MiniMax M2-her", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-24", + "last_updated": "2026-01-24", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 65532, + "input": 65532, + "output": 2048 }, "cost": { - "input": 1.15, - "output": 8, - "cache_read": 0.15 + "input": 0.30200000000000005, + "output": 1.2069999999999999 } }, - "kimi-k2.6": { - "id": "kimi-k2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "minimax/minimax-latest": { + "id": "minimax/minimax-latest", + "name": "MiniMax Latest", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "structured_output": true, - "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-05-03", + "last_updated": "2026-05-03", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 512000, + "input": 512000, + "output": 80000 }, "cost": { - "input": 0.95, - "output": 4, - "cache_read": 0.16 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06 } }, - "kimi-k2-turbo-preview": { - "id": "kimi-k2-turbo-preview", - "name": "Kimi K2 Turbo", - "family": "kimi", + "minimax/minimax-m2.7": { + "id": "minimax/minimax-m2.7", + "name": "MiniMax M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-09-05", - "last_updated": "2025-09-05", + "structured_output": true, + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 204800, + "input": 204800, + "output": 131072 }, "cost": { - "input": 2.4, - "output": 10, - "cache_read": 0.6 + "input": 0.3, + "output": 1.2 } }, - "kimi-k2-0711-preview": { - "id": "kimi-k2-0711-preview", - "name": "Kimi K2 0711", - "family": "kimi", + "kwaipilot/kat-coder-pro-v2": { + "id": "kwaipilot/kat-coder-pro-v2", + "name": "KAT Coder Pro V2", + "description": "Coding model for repository understanding, refactors, and agentic engineering tasks", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-07-14", - "last_updated": "2025-07-14", + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-28", + "last_updated": "2026-03-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 16384 + "context": 256000, + "input": 256000, + "output": 80000 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.3, + "output": 1.2 } }, - "kimi-k2-thinking": { - "id": "kimi-k2-thinking", - "name": "Kimi K2 Thinking", - "family": "kimi-thinking", + "mlabonne/NeuralDaredevil-8B-abliterated": { + "id": "mlabonne/NeuralDaredevil-8B-abliterated", + "name": "Neural Daredevil 8B abliterated", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-08", - "release_date": "2025-11-06", - "last_updated": "2025-11-06", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 8192, + "input": 8192, + "output": 8192 }, "cost": { - "input": 0.6, - "output": 2.5, - "cache_read": 0.15 + "input": 0.44, + "output": 0.44 } - } - } - }, - "berget": { - "id": "berget", - "env": ["BERGET_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.berget.ai/v1", - "name": "Berget.AI", - "doc": "https://api.berget.ai", - "models": { - "zai-org/GLM-4.7": { - "id": "zai-org/GLM-4.7", - "name": "GLM 4.7", - "family": "glm", + }, + "VongolaChouko/Starcannon-Unleashed-12B-v1.0": { + "id": "VongolaChouko/Starcannon-Unleashed-12B-v1.0", + "name": "Mistral Nemo Starcannon 12b v1", + "description": "Mistral model for multilingual chat, reasoning, and tool-assisted workflows", + "family": "mistral-nemo", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, + "context": 16384, + "input": 16384, "output": 8192 }, "cost": { - "input": 0.77, - "output": 2.75 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "mistralai/Mistral-Small-3.2-24B-Instruct-2506": { - "id": "mistralai/Mistral-Small-3.2-24B-Instruct-2506", - "name": "Mistral Small 3.2 24B Instruct 2506", - "family": "mistral-small", + "TEE/minimax-m2.5": { + "id": "TEE/minimax-m2.5", + "name": "MiniMax M2.5 TEE", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, - "temperature": true, - "knowledge": "2025-09", - "release_date": "2025-10-01", - "last_updated": "2025-10-01", + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 32000, - "output": 8192 + "context": 196608, + "input": 196608, + "output": 131072 }, "cost": { - "input": 0.33, - "output": 0.33 + "input": 0.2, + "output": 1.38 } }, - "mistralai/Mistral-Medium-3.5-128B": { - "id": "mistralai/Mistral-Medium-3.5-128B", - "name": "Mistral Medium 3.5 128B", - "family": "mistral-medium", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2026-04", - "release_date": "2026-04-29", - "last_updated": "2026-04-29", + "TEE/glm-4.7": { + "id": "TEE/glm-4.7", + "name": "GLM 4.7 TEE", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", "modalities": { - "input": ["image", "text"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 131000, + "input": 131000, + "output": 65535 }, "cost": { - "input": 1.65, - "output": 5.5 + "input": 0.85, + "output": 3.3 } }, - "meta-llama/Llama-3.3-70B-Instruct": { - "id": "meta-llama/Llama-3.3-70B-Instruct", - "name": "Llama 3.3 70B Instruct", + "TEE/llama3-3-70b": { + "id": "TEE/llama3-3-70b", + "name": "Llama 3.3 70B", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2023-12", - "release_date": "2025-04-27", - "last_updated": "2025-04-27", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-03", + "last_updated": "2025-07-03", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 8192 + "input": 128000, + "output": 16384 }, "cost": { - "input": 0.99, - "output": 0.99 + "input": 2, + "output": 2 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT-OSS-120B", - "family": "gpt-oss", + "TEE/gemma-4-31b-it": { + "id": "TEE/gemma-4-31b-it", + "name": "Gemma 4 31B IT TEE", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "structured_output": false, + "release_date": "2026-05-26", + "last_updated": "2026-05-26", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "input": 262144, + "output": 262144 }, "cost": { - "input": 0.44, - "output": 0.99 + "input": 0.15, + "output": 0.46 } }, - "google/gemma-4-31B-it": { - "id": "google/gemma-4-31B-it", - "name": "Gemma 4 31B Instruct", - "family": "gemma", - "attachment": true, - "reasoning": true, - "tool_call": true, - "structured_output": true, - "temperature": true, - "knowledge": "2025-12", - "release_date": "2026-04-02", - "last_updated": "2026-04-02", + "TEE/qwen3-30b-a3b-instruct-2507": { + "id": "TEE/qwen3-30b-a3b-instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507 TEE", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-07-29", + "last_updated": "2025-07-29", "modalities": { - "input": ["audio", "image", "text", "video"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 262000, + "input": 262000, + "output": 32768 }, "cost": { - "input": 0.275, - "output": 0.55 + "input": 0.15, + "output": 0.44999999999999996 } - } - } - }, - "github-models": { - "id": "github-models", - "env": ["GITHUB_TOKEN"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://models.github.ai/inference", - "name": "GitHub Models", - "doc": "https://docs.github.com/en/github-models", - "models": { - "deepseek/deepseek-v3-0324": { - "id": "deepseek/deepseek-v3-0324", - "name": "DeepSeek-V3-0324", - "family": "deepseek", + }, + "TEE/glm-5.1": { + "id": "TEE/glm-5.1", + "name": "GLM 5.1 TEE", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-03-24", - "last_updated": "2025-03-24", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 202752, + "input": 202752, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25, + "cache_read": 0.3 } }, - "deepseek/deepseek-r1": { - "id": "deepseek/deepseek-r1", - "name": "DeepSeek-R1", - "family": "deepseek-thinking", + "TEE/deepseek-v4-pro": { + "id": "TEE/deepseek-v4-pro", + "name": "DeepSeek V4 Pro TEE", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "structured_output": true, + "release_date": "2026-04-25", + "last_updated": "2026-04-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 8192 + "context": 800000, + "input": 800000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25, + "cache_read": 0.15 } }, - "deepseek/deepseek-r1-0528": { - "id": "deepseek/deepseek-r1-0528", - "name": "DeepSeek-R1-0528", - "family": "deepseek-thinking", + "TEE/deepseek-v4-pro:thinking": { + "id": "TEE/deepseek-v4-pro:thinking", + "name": "DeepSeek V4 Pro Thinking TEE", + "description": "Flagship DeepSeek model for coding, reasoning, and agentic work", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-05-28", - "last_updated": "2025-05-28", + "structured_output": true, + "release_date": "2026-04-29", + "last_updated": "2026-04-29", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 65536, - "output": 8192 + "context": 800000, + "input": 800000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25, + "cache_read": 0.15 } }, - "ai21-labs/ai21-jamba-1.5-mini": { - "id": "ai21-labs/ai21-jamba-1.5-mini", - "name": "AI21 Jamba 1.5 Mini", - "family": "jamba", + "TEE/gemma4-31b:thinking": { + "id": "TEE/gemma4-31b:thinking", + "name": "Gemma 4 31B Thinking TEE", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-08-29", - "last_updated": "2024-08-29", + "reasoning_options": [], + "tool_call": false, + "structured_output": true, + "release_date": "2026-05-02", + "last_updated": "2026-05-02", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.45, + "output": 1 } }, - "ai21-labs/ai21-jamba-1.5-large": { - "id": "ai21-labs/ai21-jamba-1.5-large", - "name": "AI21 Jamba 1.5 Large", - "family": "jamba", + "TEE/kimi-k2.5": { + "id": "TEE/kimi-k2.5", + "name": "Kimi K2.5 TEE", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-08-29", - "last_updated": "2024-08-29", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 256000, - "output": 4096 + "context": 128000, + "input": 128000, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.9 } }, - "microsoft/phi-3.5-mini-instruct": { - "id": "microsoft/phi-3.5-mini-instruct", - "name": "Phi-3.5-mini instruct (128k)", - "family": "phi", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "TEE/qwen2.5-vl-72b-instruct": { + "id": "TEE/qwen2.5-vl-72b-instruct", + "name": "Qwen2.5 VL 72B TEE", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-02-01", + "last_updated": "2025-02-01", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 65536, + "input": 65536, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.7, + "output": 0.7 } }, - "microsoft/phi-3-medium-4k-instruct": { - "id": "microsoft/phi-3-medium-4k-instruct", - "name": "Phi-3-medium instruct (4k)", - "family": "phi", + "TEE/gpt-oss-120b": { + "id": "TEE/gpt-oss-120b", + "name": "GPT-OSS 120B TEE", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 1024 + "context": 131072, + "input": 131072, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 2 } }, - "microsoft/phi-3.5-moe-instruct": { - "id": "microsoft/phi-3.5-moe-instruct", - "name": "Phi-3.5-MoE instruct (128k)", - "family": "phi", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "TEE/qwen3.5-27b": { + "id": "TEE/qwen3.5-27b", + "name": "Qwen3.5 27B TEE", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "attachment": true, + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-03-13", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 2.4 } }, - "microsoft/phi-3-mini-128k-instruct": { - "id": "microsoft/phi-3-mini-128k-instruct", - "name": "Phi-3-mini instruct (128k)", - "family": "phi", - "attachment": false, - "reasoning": true, + "TEE/gemma-4-26b-a4b-uncensored": { + "id": "TEE/gemma-4-26b-a4b-uncensored", + "name": "Gemma 4 26B A4B Uncensored TEE", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": true, + "release_date": "2026-05-23", + "last_updated": "2026-05-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 65536, + "input": 65536, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.7 } }, - "microsoft/phi-4-mini-instruct": { - "id": "microsoft/phi-4-mini-instruct", - "name": "Phi-4-mini-instruct", - "family": "phi", + "TEE/deepseek-v3.1": { + "id": "TEE/deepseek-v3.1", + "name": "DeepSeek V3.1 TEE", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-21", + "last_updated": "2025-08-21", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 164000, + "input": 164000, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 2.5 } }, - "microsoft/phi-4-reasoning": { - "id": "microsoft/phi-4-reasoning", - "name": "Phi-4-Reasoning", - "family": "phi", - "attachment": false, - "reasoning": true, + "TEE/kimi-k2.6": { + "id": "TEE/kimi-k2.6", + "name": "Kimi K2.6 TEE", + "description": "Kimi multimodal agent model for visual understanding, coding, and planning", + "attachment": true, + "reasoning": false, "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "structured_output": false, + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25, + "cache_read": 0.375 } }, - "microsoft/phi-3-small-8k-instruct": { - "id": "microsoft/phi-3-small-8k-instruct", - "name": "Phi-3-small instruct (8k)", - "family": "phi", + "TEE/qwen3.5-397b-a17b": { + "id": "TEE/qwen3.5-397b-a17b", + "name": "Qwen3.5 397B A17B TEE", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-28", + "last_updated": "2026-02-28", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 258048, + "input": 258048, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 3.6 } }, - "microsoft/phi-3.5-vision-instruct": { - "id": "microsoft/phi-3.5-vision-instruct", - "name": "Phi-3.5-vision instruct (128k)", - "family": "phi", + "TEE/kimi-k2.5-thinking": { + "id": "TEE/kimi-k2.5-thinking", + "name": "Kimi K2.5 Thinking TEE", + "description": "Kimi reasoning model for long-horizon research, planning, and tool use", + "family": "kimi-thinking", "attachment": false, "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-08-20", - "last_updated": "2024-08-20", + "reasoning_options": [], + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-29", + "last_updated": "2026-01-29", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 128000, - "output": 4096 + "input": 128000, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.9 } }, - "microsoft/phi-3-mini-4k-instruct": { - "id": "microsoft/phi-3-mini-4k-instruct", - "name": "Phi-3-mini instruct (4k)", - "family": "phi", - "attachment": false, + "TEE/qwen3.6-35b-a3b-uncensored": { + "id": "TEE/qwen3.6-35b-a3b-uncensored", + "name": "Qwen3.6 35B A3B Uncensored TEE", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 131072 + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "structured_output": true, + "release_date": "2026-05-23", + "last_updated": "2026-05-23", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 4096, - "output": 1024 + "context": 131072, + "input": 131072, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.5 } }, - "microsoft/phi-4-mini-reasoning": { - "id": "microsoft/phi-4-mini-reasoning", - "name": "Phi-4-mini-reasoning", - "family": "phi", + "TEE/glm-4.7-flash": { + "id": "TEE/glm-4.7-flash", + "name": "GLM 4.7 Flash TEE", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-flash", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 203000, + "input": 203000, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.5 } }, - "microsoft/phi-3-small-128k-instruct": { - "id": "microsoft/phi-3-small-128k-instruct", - "name": "Phi-3-small instruct (128k)", - "family": "phi", + "TEE/gemma4-31b": { + "id": "TEE/gemma4-31b", + "name": "Gemma 4 31B", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "reasoning": false, + "tool_call": false, + "structured_output": true, + "release_date": "2026-04-04", + "last_updated": "2026-04-04", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.45, + "output": 1 } }, - "microsoft/phi-3-medium-128k-instruct": { - "id": "microsoft/phi-3-medium-128k-instruct", - "name": "Phi-3-medium instruct (128k)", - "family": "phi", + "TEE/gemma-3-27b-it": { + "id": "TEE/gemma-3-27b-it", + "name": "Gemma 3 27B TEE", + "description": "Open Gemma instruction model for efficient chat and self-hosted deployments", + "family": "gemma", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-04-23", - "last_updated": "2024-04-23", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-03-10", + "last_updated": "2025-03-10", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 131072, + "input": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 0.8 } }, - "microsoft/phi-4": { - "id": "microsoft/phi-4", - "name": "Phi-4", - "family": "phi", + "TEE/gpt-oss-20b": { + "id": "TEE/gpt-oss-20b", + "name": "GPT-OSS 20B TEE", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 16000, - "output": 4096 + "context": 131072, + "input": 131072, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 0.8 } }, - "microsoft/phi-4-multimodal-instruct": { - "id": "microsoft/phi-4-multimodal-instruct", - "name": "Phi-4-multimodal-instruct", - "family": "phi", + "TEE/glm-5": { + "id": "TEE/glm-5", + "name": "GLM 5 TEE", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-12-11", - "last_updated": "2024-12-11", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2026-02-11", + "last_updated": "2026-02-11", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 203000, + "input": 203000, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 1.2, + "output": 3.5 } }, - "microsoft/mai-ds-r1": { - "id": "microsoft/mai-ds-r1", - "name": "MAI-DS-R1", - "family": "mai", + "TEE/glm-5.1-thinking": { + "id": "TEE/glm-5.1-thinking", + "name": "GLM 5.1 Thinking TEE", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-06", - "release_date": "2025-01-20", - "last_updated": "2025-01-20", + "structured_output": true, + "release_date": "2026-04-20", + "last_updated": "2026-04-20", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 65536, - "output": 8192 + "context": 202752, + "input": 202752, + "output": 65535 }, "cost": { - "input": 0, - "output": 0 + "input": 1.5, + "output": 5.25, + "cache_read": 0.3 } }, - "cohere/cohere-command-r-08-2024": { - "id": "cohere/cohere-command-r-08-2024", - "name": "Cohere Command R 08-2024", - "family": "command-r", + "TEE/deepseek-v3.2": { + "id": "TEE/deepseek-v3.2", + "name": "DeepSeek V3.2 TEE", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-08-01", - "last_updated": "2024-08-01", + "tool_call": false, + "structured_output": false, + "release_date": "2025-12-01", + "last_updated": "2025-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 164000, + "input": 164000, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 1 } }, - "cohere/cohere-command-a": { - "id": "cohere/cohere-command-a", - "name": "Cohere Command A", - "family": "command-a", + "TEE/qwen3.5-122b-a10b": { + "id": "TEE/qwen3.5-122b-a10b", + "name": "Qwen3.5 122B A10B TEE", + "description": "Reasoning model for deliberate analysis, multi-step problem solving, and tool use", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "structured_output": false, + "release_date": "2026-05-26", + "last_updated": "2026-05-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 262144, + "input": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.46, + "output": 3.68 } }, - "cohere/cohere-command-r-plus": { - "id": "cohere/cohere-command-r-plus", - "name": "Cohere Command R+", - "family": "command-r", + "Sao10K/L3.1-70B-Hanami-x1": { + "id": "Sao10K/L3.1-70B-Hanami-x1", + "name": "Llama 3.1 70B Hanami", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-04-04", - "last_updated": "2024-08-01", + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "cohere/cohere-command-r": { - "id": "cohere/cohere-command-r", - "name": "Cohere Command R", - "family": "command-r", + "Sao10K/L3-8B-Stheno-v3.2": { + "id": "Sao10K/L3-8B-Stheno-v3.2", + "name": "Sao10K Stheno 8b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-03-11", - "last_updated": "2024-08-01", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-11-29", + "last_updated": "2024-11-29", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 16384, + "input": 16384, + "output": 8192 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2006, + "output": 0.2006 } }, - "cohere/cohere-command-r-plus-08-2024": { - "id": "cohere/cohere-command-r-plus-08-2024", - "name": "Cohere Command R+ 08-2024", - "family": "command-r", + "Sao10K/L3.3-70B-Euryale-v2.3": { + "id": "Sao10K/L3.3-70B-Euryale-v2.3", + "name": "Llama 3.3 70B Euryale", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-08-01", - "last_updated": "2024-08-01", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 4096 + "context": 20480, + "input": 20480, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "xai/grok-3-mini": { - "id": "xai/grok-3-mini", - "name": "Grok 3 Mini", - "family": "grok", + "Sao10K/L3.1-70B-Euryale-v2.2": { + "id": "Sao10K/L3.1-70B-Euryale-v2.2", + "name": "Llama 3.1 70B Euryale", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-09", - "last_updated": "2024-12-09", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 20480, + "input": 20480, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.306, + "output": 0.357 } }, - "xai/grok-3": { - "id": "xai/grok-3", - "name": "Grok 3", - "family": "grok", + "Steelskull/L3.3-MS-Evalebis-70b": { + "id": "Steelskull/L3.3-MS-Evalebis-70b", + "name": "MS Evalebis 70b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2024-10", - "release_date": "2024-12-09", - "last_updated": "2024-12-09", + "reasoning": false, + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "openai/o1-mini": { - "id": "openai/o1-mini", - "name": "OpenAI o1-mini", - "family": "o-mini", + "Steelskull/L3.3-MS-Nevoria-70b": { + "id": "Steelskull/L3.3-MS-Nevoria-70b", + "name": "Steelskull Nevoria 70b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2023-10", - "release_date": "2024-09-12", - "last_updated": "2024-12-17", + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 65536 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "openai/gpt-4o-mini": { - "id": "openai/gpt-4o-mini", - "name": "GPT-4o mini", - "family": "gpt-mini", - "attachment": true, + "Steelskull/L3.3-Cu-Mai-R1-70b": { + "id": "Steelskull/L3.3-Cu-Mai-R1-70b", + "name": "Llama 3.3 70B Cu Mai", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", + "attachment": false, "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "tool_call": false, + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, + "context": 16384, + "input": 16384, "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "openai/o4-mini": { - "id": "openai/o4-mini", - "name": "OpenAI o4-mini", - "family": "o-mini", + "Steelskull/L3.3-MS-Evayale-70B": { + "id": "Steelskull/L3.3-MS-Evayale-70B", + "name": "Evayale 70b ", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "openai/o1-preview": { - "id": "openai/o1-preview", - "name": "OpenAI o1-preview", - "family": "o", + "Steelskull/L3.3-Nevoria-R1-70b": { + "id": "Steelskull/L3.3-Nevoria-R1-70b", + "name": "Steelskull Nevoria R1 70b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2023-10", - "release_date": "2024-09-12", - "last_updated": "2024-09-12", + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.49299999999999994, + "output": 0.49299999999999994 } }, - "openai/o1": { - "id": "openai/o1", - "name": "OpenAI o1", - "family": "o", + "Steelskull/L3.3-Electra-R1-70b": { + "id": "Steelskull/L3.3-Electra-R1-70b", + "name": "Steelskull Electra R1 70b", + "description": "Open Llama instruction model for multilingual chat, reasoning, and coding", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": false, - "temperature": false, - "knowledge": "2023-10", - "release_date": "2024-09-12", - "last_updated": "2024-12-17", + "structured_output": false, + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 200000, - "output": 100000 + "context": 16384, + "input": 16384, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.69989, + "output": 0.69989 } - }, - "openai/o3-mini": { - "id": "openai/o3-mini", - "name": "OpenAI o3-mini", - "family": "o-mini", + } + } + }, + "moark": { + "id": "moark", + "env": ["MOARK_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://moark.com/v1", + "name": "Moark", + "doc": "https://moark.com/docs/openapi/v1#tag/%E6%96%87%E6%9C%AC%E7%94%9F%E6%88%90", + "models": { + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", "attachment": false, "reasoning": true, - "tool_call": false, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 2.1, + "output": 8.4 } }, - "openai/gpt-4.1-nano": { - "id": "openai/gpt-4.1-nano", - "name": "GPT-4.1-nano", - "family": "gpt-nano", - "attachment": true, - "reasoning": false, + "GLM-4.7": { + "id": "GLM-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 3.5, + "output": 14 } - }, - "openai/o3": { - "id": "openai/o3", - "name": "OpenAI o3", - "family": "o", - "attachment": false, + } + } + }, + "lilac": { + "id": "lilac", + "env": ["LILAC_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.getlilac.com/v1", + "name": "Lilac", + "doc": "https://docs.getlilac.com/inference/models", + "models": { + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, - "tool_call": false, - "temperature": false, - "knowledge": "2024-04", - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "reasoning_options": [ + { + "type": "toggle" + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 100000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.7, + "output": 3.5, + "cache_read": 0.2 } }, - "openai/gpt-4o": { - "id": "openai/gpt-4o", - "name": "GPT-4o", - "family": "gpt", + "minimaxai/minimax-m3": { + "id": "minimaxai/minimax-m3", + "name": "MiniMax M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax-m3", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-10", - "release_date": "2024-05-13", - "last_updated": "2024-05-13", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 1048576, + "output": 1048576 }, "cost": { - "input": 0, - "output": 0 + "input": 0.28, + "output": 1.1, + "cache_read": 0.05 } }, - "openai/gpt-4.1": { - "id": "openai/gpt-4.1", - "name": "GPT-4.1", - "family": "gpt", + "google/gemma-4-31b-it": { + "id": "google/gemma-4-31b-it", + "name": "Gemma 4 31B IT", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "knowledge": "2025-01", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 262100, + "output": 262100 }, "cost": { - "input": 0, - "output": 0 + "input": 0.11, + "output": 0.35 } }, - "openai/gpt-4.1-mini": { - "id": "openai/gpt-4.1-mini", - "name": "GPT-4.1-mini", - "family": "gpt-mini", - "attachment": true, - "reasoning": false, + "zai-org/glm-5.2": { + "id": "zai-org/glm-5.2", + "name": "GLM 5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04", - "release_date": "2025-04-14", - "last_updated": "2025-04-14", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 128000, - "output": 16384 + "context": 524288, + "output": 524288 }, "cost": { - "input": 0, - "output": 0 + "input": 0.9, + "output": 3, + "cache_read": 0.27 } - }, - "meta/llama-4-scout-17b-16e-instruct": { - "id": "meta/llama-4-scout-17b-16e-instruct", - "name": "Llama 4 Scout 17B 16E Instruct", - "family": "llama", - "attachment": false, + } + } + }, + "ambient": { + "id": "ambient", + "env": ["AMBIENT_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.ambient.xyz/v1", + "name": "Ambient", + "doc": "https://ambient.xyz", + "models": { + "moonshotai/kimi-k2.7-code": { + "id": "moonshotai/kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 8192 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.84, + "output": 3.99, + "cache_read": 0.16, + "cache_write": 0 } }, - "meta/meta-llama-3.1-8b-instruct": { - "id": "meta/meta-llama-3.1-8b-instruct", - "name": "Meta-Llama-3.1-8B-Instruct", - "family": "llama", - "attachment": false, + "moonshotai/kimi-k2.6": { + "id": "moonshotai/kimi-k2.6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0, - "output": 0 + "input": 0.95, + "output": 4, + "cache_read": 0.2, + "cache_write": 0 } }, - "meta/llama-3.3-70b-instruct": { - "id": "meta/llama-3.3-70b-instruct", - "name": "Llama-3.3-70B-Instruct", - "family": "llama", + "zai-org/GLM-5.1-FP8": { + "id": "zai-org/GLM-5.1-FP8", + "name": "GLM 5.1", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 202752, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 1.4, + "output": 4.4, + "cache_read": 0, + "cache_write": 0 } }, - "meta/meta-llama-3-70b-instruct": { - "id": "meta/meta-llama-3-70b-instruct", - "name": "Meta-Llama-3-70B-Instruct", - "family": "llama", + "zai-org/GLM-5.2-FP8": { + "id": "zai-org/GLM-5.2-FP8", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 2048 + "context": 202752, + "output": 202752 }, "cost": { - "input": 0, - "output": 0 + "input": 1.2, + "output": 4.2, + "cache_read": 0.26, + "cache_write": 0 } - }, - "meta/llama-3.2-90b-vision-instruct": { - "id": "meta/llama-3.2-90b-vision-instruct", - "name": "Llama-3.2-90B-Vision-Instruct", - "family": "llama", - "attachment": false, + } + } + }, + "neon": { + "id": "neon", + "env": ["NEON_AI_GATEWAY_BASE_URL", "NEON_AI_GATEWAY_TOKEN"], + "npm": "@ai-sdk/openai-compatible", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/mlflow/v1", + "name": "Neon", + "doc": "https://neon.com/docs", + "models": { + "gemini-3-flash": { + "id": "gemini-3-flash", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 } }, - "meta/llama-3.2-11b-vision-instruct": { - "id": "meta/llama-3.2-11b-vision-instruct", - "name": "Llama-3.2-11B-Vision-Instruct", - "family": "llama", - "attachment": false, + "claude-sonnet-4": { + "id": "claude-sonnet-4", + "name": "Claude Sonnet 4.5", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-09-25", - "last_updated": "2024-09-25", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text", "image", "audio"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "meta/meta-llama-3.1-405b-instruct": { - "id": "meta/meta-llama-3.1-405b-instruct", - "name": "Meta-Llama-3.1-405B-Instruct", - "family": "llama", + "qwen3-next-80b-a3b-instruct": { + "id": "qwen3-next-80b-a3b-instruct", + "name": "Qwen3-Next 80B-A3B Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2025-04", + "release_date": "2025-09", + "last_updated": "2025-09", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, + "context": 131072, "output": 32768 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 1.2 } }, - "meta/meta-llama-3.1-70b-instruct": { - "id": "meta/meta-llama-3.1-70b-instruct", - "name": "Meta-Llama-3.1-70B-Instruct", + "llama-4-maverick": { + "id": "llama-4-maverick", + "name": "Llama 4 Maverick 17B Instruct", + "description": "Open multimodal Llama for strong reasoning with efficient everyday serving", "family": "llama", - "attachment": false, - "reasoning": true, + "attachment": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-07-23", - "last_updated": "2024-07-23", + "knowledge": "2024-08", + "release_date": "2025-04-05", + "last_updated": "2025-04-05", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 128000, - "output": 32768 + "context": 1000000, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 1.5 } }, - "meta/meta-llama-3-8b-instruct": { - "id": "meta/meta-llama-3-8b-instruct", - "name": "Meta-Llama-3-8B-Instruct", - "family": "llama", - "attachment": false, + "claude-opus-4-5": { + "id": "claude-opus-4-5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-04-18", - "last_updated": "2024-04-18", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 8192, - "output": 2048 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "meta/llama-4-maverick-17b-128e-instruct-fp8": { - "id": "meta/llama-4-maverick-17b-128e-instruct-fp8", - "name": "Llama 4 Maverick 17B 128E Instruct FP8", - "family": "llama", - "attachment": false, + "gpt-5": { + "id": "gpt-5", + "name": "GPT-5", + "description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "core42/jais-30b-chat": { - "id": "core42/jais-30b-chat", - "name": "JAIS 30b Chat", - "family": "jais", - "attachment": false, - "reasoning": true, + "gemma-3-12b": { + "id": "gemma-3-12b", + "name": "Gemma 3 12B", + "description": "Google's open-weight Gemma 3 vision-language model for text and image understanding", + "family": "gemma", + "attachment": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-03", - "release_date": "2023-08-30", - "last_updated": "2023-08-30", + "knowledge": "2024-08-31", + "release_date": "2025-03-13", + "last_updated": "2025-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 8192, - "output": 2048 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0, - "output": 0 + "input": 0.15, + "output": 0.5 } }, - "mistral-ai/mistral-nemo": { - "id": "mistral-ai/mistral-nemo", - "name": "Mistral Nemo", - "family": "mistral-nemo", - "attachment": false, + "gpt-5-1-codex-mini": { + "id": "gpt-5-1-codex-mini", + "name": "GPT-5.1 Codex mini", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-07-18", - "last_updated": "2024-07-18", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "mistral-ai/ministral-3b": { - "id": "mistral-ai/ministral-3b", - "name": "Ministral 3B", - "family": "ministral", - "attachment": false, + "gpt-5-1": { + "id": "gpt-5-1", + "name": "GPT-5.1", + "description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-03", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 128000, - "output": 8192 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0, - "output": 0 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "mistral-ai/mistral-large-2411": { - "id": "mistral-ai/mistral-large-2411", - "name": "Mistral Large 24.11", - "family": "mistral-large", + "meta-llama-3-3-70b-instruct": { + "id": "meta-llama-3-3-70b-instruct", + "name": "Llama-3.3-70B-Instruct", + "description": "Popular open Llama workhorse for multilingual chat, coding, and self-hosting", + "family": "llama", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2024-11-01", - "last_updated": "2024-11-01", + "knowledge": "2023-12", + "release_date": "2024-12-06", + "last_updated": "2024-12-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { "context": 128000, - "output": 32768 + "output": 4096 }, "cost": { - "input": 0, - "output": 0 + "input": 0.5, + "output": 1.5 } }, - "mistral-ai/mistral-small-2503": { - "id": "mistral-ai/mistral-small-2503", - "name": "Mistral Small 3.1", - "family": "mistral-small", - "attachment": false, + "gemini-3-pro": { + "id": "gemini-3-pro", + "name": "Gemini 3 Pro Preview", + "description": "Preview Gemini flagship for complex reasoning, coding, and rich multimodal prompts", + "family": "gemini-pro", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-03-01", - "last_updated": "2025-03-01", + "knowledge": "2025-01", + "release_date": "2025-11-18", + "last_updated": "2025-11-18", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "mistral-ai/mistral-medium-2505": { - "id": "mistral-ai/mistral-medium-2505", - "name": "Mistral Medium 3 (25.05)", - "family": "mistral-medium", - "attachment": false, + "gpt-5-2": { + "id": "gpt-5-2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2024-09", - "release_date": "2025-05-01", - "last_updated": "2025-05-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 32768 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0, - "output": 0 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "mistral-ai/codestral-2501": { - "id": "mistral-ai/codestral-2501", - "name": "Codestral 25.01", - "family": "codestral", - "attachment": false, + "gemini-3-1-flash-lite": { + "id": "gemini-3-1-flash-lite", + "name": "Gemini 3.1 Flash Lite Preview", + "description": "Low-latency Gemini model for high-volume multimodal and agent workloads", + "family": "gemini-flash-lite", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-03", - "release_date": "2025-01-01", - "last_updated": "2025-01-01", + "knowledge": "2025-01", + "release_date": "2026-03-03", + "last_updated": "2026-03-03", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32000, - "output": 8192 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0, - "output": 0 + "input": 0.25, + "output": 1.5, + "cache_read": 0.025, + "input_audio": 0.5 } - } - } - }, - "neuralwatt": { - "id": "neuralwatt", - "env": ["NEURALWATT_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.neuralwatt.com/v1", - "name": "Neuralwatt", - "doc": "https://portal.neuralwatt.com/docs", - "models": { - "glm-5-fast": { - "id": "glm-5-fast", - "name": "GLM 5 Fast", - "family": "glm", - "attachment": false, - "reasoning": false, + }, + "claude-sonnet-4-5": { + "id": "claude-sonnet-4-5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 200000, - "output": 200000 + "output": 64000 }, "cost": { - "input": 1.1, - "output": 3.6 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "kimi-k2.6-fast": { - "id": "kimi-k2.6-fast", - "name": "Kimi K2.6 Fast", - "family": "kimi", + "claude-opus-4-7": { + "id": "claude-opus-4-7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0.69, - "output": 3.22 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen3.5-397b-fast": { - "id": "qwen3.5-397b-fast", - "name": "Qwen3.5 397B Fast", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gemini-2-5-pro": { + "id": "gemini-2-5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.69, - "output": 4.14 + "input": 1.25, + "output": 10, + "cache_read": 0.125, + "tiers": [ + { + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2.5, + "output": 15, + "cache_read": 0.25 + } } }, - "glm-5.1-fast": { - "id": "glm-5.1-fast", - "name": "GLM 5.1 Fast", - "family": "glm", - "attachment": false, - "reasoning": false, + "claude-opus-4-8": { + "id": "claude-opus-4-8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 200000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 1.1, - "output": 3.6 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "qwen3.6-35b-fast": { - "id": "qwen3.6-35b-fast", - "name": "Qwen3.6 35B Fast", - "family": "qwen3.6", + "gpt-5-3-codex": { + "id": "gpt-5-3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.05, - "output": 0.1 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "kimi-k2.5-fast": { - "id": "kimi-k2.5-fast", - "name": "Kimi K2.5 Fast", - "family": "kimi", + "gpt-5-4-nano": { + "id": "gpt-5-4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", + "family": "gpt-nano", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.52, - "output": 2.59 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "Qwen/Qwen3.5-397B-A17B-FP8": { - "id": "Qwen/Qwen3.5-397B-A17B-FP8", - "name": "Qwen3.5 397B A17B FP8", - "family": "qwen", - "attachment": false, + "claude-opus-4-1": { + "id": "claude-opus-4-1", + "name": "Claude Opus 4.1 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 31999 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "knowledge": "2025-03-31", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 200000, + "output": 32000 }, "cost": { - "input": 0.69, - "output": 4.14 + "input": 15, + "output": 75, + "cache_read": 1.5, + "cache_write": 18.75 } }, - "Qwen/Qwen3.6-35B-A3B": { - "id": "Qwen/Qwen3.6-35B-A3B", - "name": "Qwen3.6 35B A3B", - "family": "qwen3.6", + "gpt-5-4-mini": { + "id": "gpt-5-4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-17", + "last_updated": "2026-03-17", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.05, - "output": 0.1 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "zai-org/GLM-5.1-FP8": { - "id": "zai-org/GLM-5.1-FP8", - "name": "GLM 5.1 FP8", - "family": "glm", - "attachment": false, + "gemini-2-5-flash": { + "id": "gemini-2-5-flash", + "name": "Gemini 2.5 Flash", + "description": "Fast Gemini workhorse for multimodal apps where latency and price matter", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 0, + "max": 24576 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 200000, - "output": 200000 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 1.1, - "output": 3.6 + "input": 0.3, + "output": 2.5, + "cache_read": 0.03, + "input_audio": 1 } }, - "mistralai/Devstral-Small-2-24B-Instruct-2512": { - "id": "mistralai/Devstral-Small-2-24B-Instruct-2512", - "name": "Devstral Small 2 24B Instruct 2512", - "family": "devstral", + "gpt-5-2-codex": { + "id": "gpt-5-2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", "attachment": true, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2025-12-09", - "last_updated": "2025-12-09", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.12, - "output": 0.35 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "openai/gpt-oss-20b": { - "id": "openai/gpt-oss-20b", - "name": "GPT OSS 20B", + "gpt-oss-120b": { + "id": "gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open GPT reasoning model for self-hosted agents and controllable deployments", "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, "release_date": "2025-08-05", "last_updated": "2025-08-05", @@ -113441,1887 +167977,2287 @@ }, "open_weights": true, "limit": { - "context": 16384, - "output": 16384 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.03, - "output": 0.16 + "input": 0.072, + "output": 0.28 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "Kimi K2.6", - "family": "kimi", + "gemini-3-5-flash": { + "id": "gemini-3-5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.69, - "output": 3.22 + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", + "gemini-3-1-pro": { + "id": "gemini-3-1-pro", + "name": "Gemini 3.1 Pro Preview Custom Tools", + "description": "Advanced Gemini model for complex reasoning, coding, and multimodal analysis", + "family": "gemini-pro", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-01-27", - "last_updated": "2026-01-27", + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video", "audio", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1048576, + "output": 65536 }, "cost": { - "input": 0.52, - "output": 2.59 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax M2.5", - "family": "minimax", - "attachment": false, + "claude-haiku-4-5": { + "id": "claude-haiku-4-5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 196608, - "output": 196608 + "context": 200000, + "output": 64000 }, "cost": { - "input": 0.35, - "output": 1.38 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } - } - } - }, - "sarvam": { - "id": "sarvam", - "env": ["SARVAM_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.sarvam.ai/v1", - "name": "Sarvam AI", - "doc": "https://docs.sarvam.ai/api-reference-docs/getting-started/models", - "models": { - "sarvam-105b": { - "id": "sarvam-105b", - "name": "Sarvam-105B", - "family": "sarvam", - "attachment": false, + }, + "claude-opus-4-6": { + "id": "claude-opus-4-6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", + "family": "claude-opus", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 127999 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "release_date": "2026-02-18", - "last_updated": "2026-03-06", + "knowledge": "2025-05-31", + "release_date": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "sarvam-30b": { - "id": "sarvam-30b", - "name": "Sarvam-30B", - "family": "sarvam", + "qwen35-122b-a10b": { + "id": "qwen35-122b-a10b", + "name": "Qwen3.5 122B-A10B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-18", - "last_updated": "2026-03-06", + "release_date": "2026-02-23", + "last_updated": "2026-02-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 65536, - "output": 65536 + "context": 262144, + "output": 8000 + }, + "cost": { + "input": 0.22, + "output": 2.2 } - } - } - }, - "togetherai": { - "id": "togetherai", - "env": ["TOGETHER_API_KEY"], - "npm": "@ai-sdk/togetherai", - "name": "Together AI", - "doc": "https://docs.together.ai/docs/serverless-models", - "models": { - "essentialai/Rnj-1-Instruct": { - "id": "essentialai/Rnj-1-Instruct", - "name": "Rnj-1 Instruct", - "family": "rnj", + }, + "meta-llama-3-1-8b-instruct": { + "id": "meta-llama-3-1-8b-instruct", + "name": "Llama 3.1 8B Instruct", + "description": "Meta's compact open-weight Llama 3.1 model for fast, low-cost text generation", + "family": "llama", "attachment": false, "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12-05", - "last_updated": "2025-12-05", + "knowledge": "2023-12-31", + "release_date": "2024-07-23", + "last_updated": "2024-07-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 32768 + "context": 131072, + "output": 16384 }, "cost": { "input": 0.15, - "output": 0.15 + "output": 0.45 } }, - "Qwen/Qwen3.5-397B-A17B": { - "id": "Qwen/Qwen3.5-397B-A17B", - "name": "Qwen3.5 397B A17B", - "family": "qwen", - "attachment": false, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-02-16", - "last_updated": "2026-02-16", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { "input": ["text", "image"], - "output": ["text"] + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 130000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.6, - "output": 3.6 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "Qwen/Qwen3.6-Plus": { - "id": "Qwen/Qwen3.6-Plus", - "name": "Qwen3.6 Plus", - "family": "qwen", - "attachment": false, + "gpt-5-nano": { + "id": "gpt-5-nano", + "name": "GPT-5 Nano", + "description": "Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs", + "family": "gpt-nano", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + } + ], "tool_call": true, - "temperature": true, - "release_date": "2026-04-30", - "last_updated": "2026-04-30", + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 500000 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.5, - "output": 3 + "input": 0.05, + "output": 0.4, + "cache_read": 0.005 } }, - "Qwen/Qwen3-Coder-Next-FP8": { - "id": "Qwen/Qwen3-Coder-Next-FP8", - "name": "Qwen3 Coder Next FP8", - "family": "qwen", - "attachment": false, + "claude-sonnet-4-6": { + "id": "claude-sonnet-4-6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 63999 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2026-02-03", - "release_date": "2026-02-03", - "last_updated": "2026-02-03", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1000000, + "output": 64000 }, "cost": { - "input": 0.5, - "output": 1.2 + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507-tput", - "name": "Qwen3 235B A22B Instruct 2507 FP8", - "family": "qwen", - "attachment": false, + "gpt-5-4": { + "id": "gpt-5-4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image", "pdf"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 1050000, + "input": 922000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 0.2, - "output": 0.6 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": { - "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8", - "name": "Qwen3 Coder 480B A35B Instruct", - "family": "qwen", - "attachment": false, - "reasoning": false, + "gpt-5-1-codex-max": { + "id": "gpt-5-1-codex-max", + "name": "GPT-5.1 Codex Max", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", + "family": "gpt-codex", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-23", - "last_updated": "2025-07-23", + "structured_output": true, + "temperature": false, + "knowledge": "2024-09-30", + "release_date": "2025-11-13", + "last_updated": "2025-11-13", "modalities": { - "input": ["text"], - "output": ["text"] + "input": ["text", "image"], + "output": ["text", "image"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 400000, + "input": 272000, + "output": 128000 + }, + "provider": { + "npm": "@ai-sdk/openai", + "api": "${NEON_AI_GATEWAY_BASE_URL}/ai-gateway/openai/v1", + "shape": "responses" }, "cost": { - "input": 2, - "output": 2 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "zai-org/GLM-5.1": { - "id": "zai-org/GLM-5.1", - "name": "GLM-5.1", - "family": "glm", + "gpt-oss-20b": { + "id": "gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 131072 + "context": 131072, + "output": 32768 }, "cost": { - "input": 1.4, - "output": 4.4 + "input": 0.05, + "output": 0.2 } - }, - "meta-llama/Llama-3.3-70B-Instruct-Turbo": { - "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo", - "name": "Llama 3.3 70B", - "family": "llama", + } + } + }, + "upstage": { + "id": "upstage", + "env": ["UPSTAGE_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.upstage.ai/v1/solar", + "name": "Upstage", + "doc": "https://developers.upstage.ai/docs/apis/chat", + "models": { + "solar-pro2": { + "id": "solar-pro2", + "name": "solar-pro2", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "solar-pro", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2023-12", - "release_date": "2024-12-06", - "last_updated": "2024-12-06", + "knowledge": "2025-03", + "release_date": "2025-05-20", + "last_updated": "2025-05-20", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, - "output": 131072 + "context": 65536, + "output": 8192 }, "cost": { - "input": 0.88, - "output": 0.88 + "input": 0.25, + "output": 0.25 } }, - "deepseek-ai/DeepSeek-V3": { - "id": "deepseek-ai/DeepSeek-V3", - "name": "DeepSeek V3", - "family": "deepseek", + "solar-pro3": { + "id": "solar-pro3", + "name": "solar-pro3", + "description": "Flagship model for demanding analysis, coding, and production agent workflows", + "family": "solar-pro", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2025-01-20", - "last_updated": "2025-05-29", + "knowledge": "2025-03", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { "context": 131072, - "output": 131072 + "output": 8192 }, "cost": { - "input": 1.25, - "output": 1.25 + "input": 0.25, + "output": 0.25 } }, - "deepseek-ai/DeepSeek-R1": { - "id": "deepseek-ai/DeepSeek-R1", - "name": "DeepSeek R1", - "family": "deepseek-thinking", + "solar-mini": { + "id": "solar-mini", + "name": "solar-mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "solar-mini", "attachment": false, - "reasoning": true, - "tool_call": false, + "reasoning": false, + "tool_call": true, "temperature": true, - "knowledge": "2024-07", - "release_date": "2024-12-26", - "last_updated": "2025-03-24", + "knowledge": "2024-09", + "release_date": "2024-06-12", + "last_updated": "2025-04-22", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 163839, - "output": 163839 + "context": 32768, + "output": 4096 }, "cost": { - "input": 3, - "output": 7 + "input": 0.15, + "output": 0.15 } - }, - "deepseek-ai/DeepSeek-V3-1": { - "id": "deepseek-ai/DeepSeek-V3-1", - "name": "DeepSeek V3.1", - "family": "deepseek", + } + } + }, + "zhipuai-coding-plan": { + "id": "zhipuai-coding-plan", + "env": ["ZHIPU_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://open.bigmodel.cn/api/coding/paas/v4", + "name": "Zhipu AI Coding Plan", + "doc": "https://docs.bigmodel.cn/cn/coding-plan/overview", + "models": { + "glm-5.1": { + "id": "glm-5.1", + "name": "GLM-5.1", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-21", - "last_updated": "2025-08-21", + "release_date": "2026-03-27", + "last_updated": "2026-03-27", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 200000, "output": 131072 }, "cost": { - "input": 0.6, - "output": 1.7 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "deepseek-ai/DeepSeek-V4-Pro": { - "id": "deepseek-ai/DeepSeek-V4-Pro", - "name": "DeepSeek V4 Pro", - "family": "deepseek", - "attachment": false, + "glm-5v-turbo": { + "id": "glm-5v-turbo", + "name": "GLM-5V-Turbo", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, - "structured_output": true, "temperature": true, - "release_date": "2026-04-24", - "last_updated": "2026-04-24", + "release_date": "2026-04-01", + "last_updated": "2026-04-01", "modalities": { - "input": ["text"], + "input": ["text", "image", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 512000, - "output": 384000 + "context": 200000, + "output": 131072 }, "cost": { - "input": 2.1, - "output": 4.4, - "cache_read": 0.2 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "openai/gpt-oss-120b": { - "id": "openai/gpt-oss-120b", - "name": "GPT OSS 120B", - "family": "gpt-oss", + "glm-5-turbo": { + "id": "glm-5-turbo", + "name": "GLM-5-Turbo", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-08", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-03-16", + "last_updated": "2026-03-16", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 131072, + "context": 200000, "output": 131072 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "google/gemma-4-31B-it": { - "id": "google/gemma-4-31B-it", - "name": "Gemma 4 31B Instruct", - "family": "gemma", - "attachment": true, + "glm-4.5-air": { + "id": "glm-4.5-air", + "name": "GLM-4.5-Air", + "description": "Efficient GLM model for fast reasoning, coding, and agent workflows", + "family": "glm-air", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-07", - "last_updated": "2026-04-07", + "knowledge": "2025-04", + "release_date": "2025-07-28", + "last_updated": "2025-07-28", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 131072 + "context": 131072, + "output": 98304 }, "cost": { - "input": 0.2, - "output": 0.5 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "moonshotai/Kimi-K2.6": { - "id": "moonshotai/Kimi-K2.6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "glm-4.6v": { + "id": "glm-4.6v", + "name": "GLM-4.6V", + "description": "GLM vision model for visual reasoning, documents, and multimodal agents", + "family": "glm", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "knowledge": "2025-04", + "release_date": "2025-12-08", + "last_updated": "2025-12-08", "modalities": { "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 131000 + "context": 128000, + "output": 32768 }, "cost": { - "input": 1.2, - "output": 4.5, - "cache_read": 0.2 + "input": 0.3, + "output": 0.9 } }, - "moonshotai/Kimi-K2.5": { - "id": "moonshotai/Kimi-K2.5", - "name": "Kimi K2.5", - "family": "kimi", + "glm-5.2": { + "id": "glm-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, - "interleaved": true, - "temperature": true, - "knowledge": "2026-01", - "release_date": "2026-01-27", - "last_updated": "2026-01-27", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 262144, - "output": 262144 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.5, - "output": 2.8 - } - }, - "MiniMaxAI/MiniMax-M2.5": { - "id": "MiniMaxAI/MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", - "attachment": false, - "reasoning": true, - "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, + "context": 1000000, "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "MiniMaxAI/MiniMax-M2.7": { - "id": "MiniMaxAI/MiniMax-M2.7", - "name": "MiniMax-M2.7", - "family": "minimax", + "glm-4.7": { + "id": "glm-4.7", + "name": "GLM-4.7", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "structured_output": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-04", + "release_date": "2025-12-22", + "last_updated": "2025-12-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, + "context": 204800, "output": 131072 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.06 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } } } }, - "qihang-ai": { - "id": "qihang-ai", - "env": ["QIHANG_API_KEY"], + "chutes": { + "id": "chutes", + "env": ["CHUTES_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.qhaigc.net/v1", - "name": "QiHang", - "doc": "https://www.qhaigc.net/docs", + "api": "https://llm.chutes.ai/v1", + "name": "Chutes", + "doc": "https://llm.chutes.ai/v1/models", "models": { - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", + "moonshotai/Kimi-K2.6-TEE": { + "id": "moonshotai/Kimi-K2.6-TEE", + "name": "Kimi K2.6 TEE", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", - "modalities": { - "input": ["text", "image"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 32000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.71, - "output": 3.57 - } - }, - "gemini-3-flash-preview": { - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash Preview", - "family": "gemini-flash", - "attachment": true, - "reasoning": true, - "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "knowledge": "2025-12", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, - "output": 65536 + "context": 262144, + "output": 65535 }, "cost": { - "input": 0.07, - "output": 0.43, - "tiers": [ - { - "input": 0.07, - "output": 0.43, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 0.07, - "output": 0.43 - } + "input": 0.66, + "output": 3.5, + "cache_read": 0.33 } }, - "gpt-5-mini": { - "id": "gpt-5-mini", - "name": "GPT-5-Mini", - "family": "gpt-mini", + "moonshotai/Kimi-K2.5-TEE": { + "id": "moonshotai/Kimi-K2.5-TEE", + "name": "Kimi K2.5 TEE", + "description": "Earlier Kimi frontier model for long-context agents, coding, and multimodal work", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-09-30", - "release_date": "2025-09-15", - "last_updated": "2025-09-15", + "knowledge": "2024-10", + "release_date": "2026-01", + "last_updated": "2026-01", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 262144, + "output": 65535 }, "cost": { - "input": 0.04, - "output": 0.29 + "input": 0.44, + "output": 2, + "cache_read": 0.22 } }, - "gemini-3-pro-preview": { - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro Preview", - "family": "gemini-pro", + "google/gemma-4-31B-turbo-TEE": { + "id": "google/gemma-4-31B-turbo-TEE", + "name": "gemma 4 31B turbo TEE", + "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning", + "family": "gemma", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-11", - "release_date": "2025-11-19", - "last_updated": "2025-11-19", + "release_date": "2026-04-02", + "last_updated": "2026-04-02", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 65000 + "context": 131072, + "output": 65536 }, "cost": { - "input": 0.57, - "output": 3.43 + "input": 0.12, + "output": 0.37, + "cache_read": 0.06 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "Qwen/Qwen3-32B-TEE": { + "id": "Qwen/Qwen3-32B-TEE", + "name": "Qwen3 32B TEE", + "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "knowledge": "2025-04", + "release_date": "2025-04", + "last_updated": "2025-04", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 40960, + "output": 40960 }, "cost": { - "input": 0.43, - "output": 2.14 + "input": 0.104, + "output": 0.416, + "cache_read": 0.052 } }, - "gpt-5.2": { - "id": "gpt-5.2", - "name": "GPT-5.2", - "family": "gpt", + "Qwen/Qwen3.6-27B-TEE": { + "id": "Qwen/Qwen3.6-27B-TEE", + "name": "Qwen3.6 27B TEE", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.25, - "output": 2 + "input": 0.3, + "output": 2, + "cache_read": 0.15 } }, - "gpt-5.2-codex": { - "id": "gpt-5.2-codex", - "name": "GPT-5.2 Codex", - "family": "gpt-codex", - "attachment": true, + "Qwen/Qwen3-235B-A22B-Thinking-2507-TEE": { + "id": "Qwen/Qwen3-235B-A22B-Thinking-2507-TEE", + "name": "Qwen3 235B A22B Thinking 2507 TEE", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, - "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2025-12-11", - "last_updated": "2025-12-11", + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07", + "last_updated": "2026-06-21", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 262144, + "output": 262144 }, "cost": { - "input": 0.14, - "output": 1.14 + "input": 0.2989, + "output": 1.1957, + "cache_read": 0.14945 } }, - "gemini-2.5-flash": { - "id": "gemini-2.5-flash", - "name": "Gemini 2.5 Flash", - "family": "gemini-flash", + "Qwen/Qwen3.5-397B-A17B-TEE": { + "id": "Qwen/Qwen3.5-397B-A17B-TEE", + "name": "Qwen3.5 397B A17B TEE", + "description": "Large open Qwen multimodal MoE for visual agents and long technical tasks", + "family": "qwen", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "knowledge": "2025-01", - "release_date": "2025-12-17", - "last_updated": "2025-12-17", + "release_date": "2026-02-15", + "last_updated": "2026-02-15", "modalities": { - "input": ["text", "image", "video", "audio", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, + "context": 262144, "output": 65536 }, "cost": { - "input": 0.09, - "output": 0.71, - "tiers": [ - { - "input": 0.09, - "output": 0.71, - "tier": { - "type": "context", - "size": 200000 - } - } - ], - "context_over_200k": { - "input": 0.09, - "output": 0.71 - } + "input": 0.45, + "output": 3, + "cache_read": 0.225 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", - "attachment": true, - "reasoning": true, - "tool_call": true, + "unsloth/Mistral-Nemo-Instruct-2407-TEE": { + "id": "unsloth/Mistral-Nemo-Instruct-2407-TEE", + "name": "Mistral Nemo Instruct 2407 TEE", + "description": "Efficient Mistral-NVIDIA open model for multilingual chat and local deployment", + "family": "mistral-nemo", + "attachment": false, + "reasoning": false, + "tool_call": false, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-10-01", - "last_updated": "2025-10-01", + "knowledge": "2024-07", + "release_date": "2024-07-01", + "last_updated": "2024-07-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 131072, + "output": 131072 }, "cost": { - "input": 0.14, - "output": 0.71 + "input": 0.0245, + "output": 0.0978, + "cache_read": 0.01225 } - } - } - }, - "tencent-tokenhub": { - "id": "tencent-tokenhub", - "env": ["TENCENT_TOKENHUB_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://tokenhub.tencentmaas.com/v1", - "name": "Tencent TokenHub", - "doc": "https://cloud.tencent.com/document/product/1823/130050", - "models": { - "hy3-preview": { - "id": "hy3-preview", - "name": "Hy3 preview", - "family": "Hy", + }, + "zai-org/GLM-5-TEE": { + "id": "zai-org/GLM-5-TEE", + "name": "GLM 5 TEE", + "description": "General GLM flagship for coding, analysis, and tool-heavy engineering workflows", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-04-20", - "last_updated": "2026-04-20", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 256000, - "output": 64000 - }, - "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 - } - } - } - }, - "anthropic": { - "id": "anthropic", - "env": ["ANTHROPIC_API_KEY"], - "npm": "@ai-sdk/anthropic", - "name": "Anthropic", - "doc": "https://docs.anthropic.com/en/docs/about-claude/models", - "models": { - "claude-3-sonnet-20240229": { - "id": "claude-3-sonnet-20240229", - "name": "Claude Sonnet 3", - "family": "claude-sonnet", - "attachment": true, - "reasoning": false, - "tool_call": true, - "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-04", - "last_updated": "2024-03-04", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 4096 + "context": 202752, + "output": 65535 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 0.3 + "input": 0.95, + "output": 2.55, + "cache_read": 0.475 } }, - "claude-haiku-4-5": { - "id": "claude-haiku-4-5", - "name": "Claude Haiku 4.5 (latest)", - "family": "claude-haiku", - "attachment": true, + "zai-org/GLM-5.1-TEE": { + "id": "zai-org/GLM-5.1-TEE", + "name": "GLM 5.1 TEE", + "description": "Strong GLM coding model for agentic engineering, terminals, and repository generation", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "release_date": "2026-04-07", + "last_updated": "2026-04-07", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 202752, + "output": 65535 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.98, + "output": 3.08, + "cache_read": 0.49 } }, - "claude-opus-4-5-20251101": { - "id": "claude-opus-4-5-20251101", - "name": "Claude Opus 4.5", - "family": "claude-opus", - "attachment": true, + "zai-org/GLM-5.2-TEE": { + "id": "zai-org/GLM-5.2-TEE", + "name": "GLM 5.2 TEE", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-01", - "last_updated": "2025-11-01", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 64000 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 - } - }, - "claude-3-opus-20240229": { - "id": "claude-3-opus-20240229", - "name": "Claude Opus 3", - "family": "claude-opus", - "attachment": true, - "reasoning": false, - "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-02-29", - "last_updated": "2024-02-29", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 4096 + "context": 1048576, + "output": 65535 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1.4, + "output": 4.4, + "cache_read": 0.7 } }, - "claude-3-5-haiku-20241022": { - "id": "claude-3-5-haiku-20241022", - "name": "Claude Haiku 3.5", - "family": "claude-haiku", - "attachment": true, - "reasoning": false, + "deepseek-ai/DeepSeek-V3.2-TEE": { + "id": "deepseek-ai/DeepSeek-V3.2-TEE", + "name": "DeepSeek V3.2 TEE", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", - "modalities": { - "input": ["text", "image", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 8192 + "interleaved": { + "field": "reasoning_content" }, - "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 - } - }, - "claude-3-5-sonnet-20241022": { - "id": "claude-3-5-sonnet-20241022", - "name": "Claude Sonnet 3.5 v2", - "family": "claude-sonnet", - "attachment": true, - "reasoning": false, - "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2025-12", + "last_updated": "2026-06-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 131072, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1, + "output": 1, + "cache_read": 0.5 } }, - "claude-sonnet-4-6": { - "id": "claude-sonnet-4-6", - "name": "Claude Sonnet 4.6", - "family": "claude-sonnet", - "attachment": true, + "MiniMaxAI/MiniMax-M2.5-TEE": { + "id": "MiniMaxAI/MiniMax-M2.5-TEE", + "name": "MiniMax M2.5 TEE", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-03-13", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 64000 + "context": 196608, + "output": 65536 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.15, + "output": 1.2, + "cache_read": 0.075 } - }, - "claude-opus-4-0": { - "id": "claude-opus-4-0", - "name": "Claude Opus 4 (latest)", - "family": "claude-opus", - "attachment": true, + } + } + }, + "minimax-cn-coding-plan": { + "id": "minimax-cn-coding-plan", + "env": ["MINIMAX_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.minimaxi.com/anthropic/v1", + "name": "MiniMax Token Plan (minimaxi.com)", + "doc": "https://platform.minimaxi.com/docs/token-plan/intro", + "models": { + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0 } }, - "claude-opus-4-7": { - "id": "claude-opus-4-7", - "name": "Claude Opus 4.7", - "family": "claude-opus", - "attachment": true, + "MiniMax-M2.5-highspeed": { + "id": "MiniMax-M2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "temperature": true, + "release_date": "2026-02-13", + "last_updated": "2026-02-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 30, - "output": 150, - "cache_read": 3, - "cache_write": 37.5 - }, - "provider": { - "body": { - "speed": "fast" - }, - "headers": { - "anthropic-beta": "fast-mode-2026-02-01" - } - } - } - } + "context": 204800, + "output": 131072 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "claude-3-haiku-20240307": { - "id": "claude-3-haiku-20240307", - "name": "Claude Haiku 3", - "family": "claude-haiku", - "attachment": true, - "reasoning": false, + "MiniMax-M2.7-highspeed": { + "id": "MiniMax-M2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2023-08-31", - "release_date": "2024-03-13", - "last_updated": "2024-03-13", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 4096 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.25, - "output": 1.25, - "cache_read": 0.03, - "cache_write": 0.3 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "claude-sonnet-4-5-20250929": { - "id": "claude-sonnet-4-5-20250929", - "name": "Claude Sonnet 4.5", - "family": "claude-sonnet", - "attachment": true, + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax-M2", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 196608, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0, + "output": 0 } }, - "claude-3-5-haiku-latest": { - "id": "claude-3-5-haiku-latest", - "name": "Claude Haiku 3.5 (latest)", - "family": "claude-haiku", - "attachment": true, - "reasoning": false, + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2024-07-31", - "release_date": "2024-10-22", - "last_updated": "2024-10-22", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0.8, - "output": 4, - "cache_read": 0.08, - "cache_write": 1 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "claude-opus-4-1": { - "id": "claude-opus-4-1", - "name": "Claude Opus 4.1 (latest)", - "family": "claude-opus", + "MiniMax-M3": { + "id": "MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal coding model for long-context reasoning and agent tasks", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "release_date": "2026-06-01", + "last_updated": "2026-06-25", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } }, - "claude-sonnet-4-0": { - "id": "claude-sonnet-4-0", - "name": "Claude Sonnet 4 (latest)", - "family": "claude-sonnet", - "attachment": true, + "MiniMax-M2.7": { + "id": "MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 204800, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0, + "output": 0, + "cache_read": 0, + "cache_write": 0 } - }, - "claude-3-5-sonnet-20240620": { - "id": "claude-3-5-sonnet-20240620", - "name": "Claude Sonnet 3.5", - "family": "claude-sonnet", - "attachment": true, - "reasoning": false, + } + } + }, + "deepseek": { + "id": "deepseek", + "env": ["DEEPSEEK_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.deepseek.com", + "name": "DeepSeek", + "doc": "https://api-docs.deepseek.com/quick_start/pricing", + "models": { + "deepseek-v4-flash": { + "id": "deepseek-v4-flash", + "name": "DeepSeek V4 Flash", + "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work", + "family": "deepseek-flash", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2024-04-30", - "release_date": "2024-06-20", - "last_updated": "2024-06-20", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 8192 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "claude-opus-4-5": { - "id": "claude-opus-4-5", - "name": "Claude Opus 4.5 (latest)", - "family": "claude-opus", - "attachment": true, + "deepseek-v4-pro": { + "id": "deepseek-v4-pro", + "name": "DeepSeek V4 Pro", + "description": "Open MoE flagship with million-token context for coding and long agent runs", + "family": "deepseek-thinking", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-11-24", - "last_updated": "2025-11-24", + "knowledge": "2025-05", + "release_date": "2026-04-24", + "last_updated": "2026-04-24", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 0.435, + "output": 0.87, + "cache_read": 0.003625 } }, - "claude-opus-4-1-20250805": { - "id": "claude-opus-4-1-20250805", - "name": "Claude Opus 4.1", - "family": "claude-opus", + "deepseek-reasoner": { + "id": "deepseek-reasoner", + "name": "DeepSeek Reasoner", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } }, - "claude-haiku-4-5-20251001": { - "id": "claude-haiku-4-5-20251001", - "name": "Claude Haiku 4.5", - "family": "claude-haiku", + "deepseek-chat": { + "id": "deepseek-chat", + "name": "DeepSeek Chat", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2025-10-15", - "last_updated": "2025-10-15", + "knowledge": "2025-09", + "release_date": "2025-12-01", + "last_updated": "2026-02-28", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1000000, + "output": 384000 }, "cost": { - "input": 1, - "output": 5, - "cache_read": 0.1, - "cache_write": 1.25 + "input": 0.14, + "output": 0.28, + "cache_read": 0.0028 } - }, - "claude-sonnet-4-20250514": { - "id": "claude-sonnet-4-20250514", - "name": "Claude Sonnet 4", - "family": "claude-sonnet", - "attachment": true, + } + } + }, + "wafer.ai": { + "id": "wafer.ai", + "env": ["WAFER_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://pass.wafer.ai/v1", + "name": "Wafer", + "doc": "https://docs.wafer.ai/wafer-pass", + "models": { + "glm5.2-fast": { + "id": "glm5.2-fast", + "name": "GLM5.2-Fast", + "description": "The same model served for high TPS.", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1048576, + "output": 131072 }, "cost": { "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "output": 10.25, + "cache_read": 0.5, + "cache_write": 0 } }, - "claude-opus-4-6": { - "id": "claude-opus-4-6", - "name": "Claude Opus 4.6", - "family": "claude-opus", + "Kimi-K2.6": { + "id": "Kimi-K2.6", + "name": "Kimi K2.6", + "description": "Kimi K2.6 sparse MoE model with a 262K context window. Available serverless and not included in standard Wafer Pass. Non-ZDR only: requests with `Wafer-ZDR: required` are rejected.", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-05-31", - "release_date": "2026-02-05", - "last_updated": "2026-03-13", + "structured_output": true, + "temperature": false, + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image", "video"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1000000, - "output": 128000 - }, - "experimental": { - "modes": { - "fast": { - "cost": { - "input": 30, - "output": 150, - "cache_read": 3, - "cache_write": 37.5 - }, - "provider": { - "body": { - "speed": "fast" - }, - "headers": { - "anthropic-beta": "fast-mode-2026-02-01" - } - } - } - } + "context": 262144, + "output": 65536 }, "cost": { - "input": 5, - "output": 25, - "cache_read": 0.5, - "cache_write": 6.25 + "input": 1.14, + "output": 4.8, + "cache_read": 0.19, + "cache_write": 0 } }, - "claude-3-7-sonnet-20250219": { - "id": "claude-3-7-sonnet-20250219", - "name": "Claude Sonnet 3.7", - "family": "claude-sonnet", + "MiniMax-M3": { + "id": "MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "max"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2024-10-31", - "release_date": "2025-02-19", - "last_updated": "2025-02-19", + "release_date": "2026-06-01", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text", "image"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1048576, + "output": 128000 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 0.33, + "output": 1.32, + "cache_read": 0.07, + "cache_write": 0, + "tiers": [ + { + "input": 0.66, + "output": 2.64, + "cache_read": 0.13, + "cache_write": 0, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.66, + "output": 2.64, + "cache_read": 0.13, + "cache_write": 0 + } } }, - "claude-sonnet-4-5": { - "id": "claude-sonnet-4-5", - "name": "Claude Sonnet 4.5 (latest)", - "family": "claude-sonnet", - "attachment": true, + "GLM-5.2": { + "id": "GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2025-09-29", - "last_updated": "2025-09-29", + "release_date": "2026-06-13", + "last_updated": "2026-06-22", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 64000 + "context": 1048576, + "output": 131072 }, "cost": { - "input": 3, - "output": 15, - "cache_read": 0.3, - "cache_write": 3.75 + "input": 1.2, + "output": 4.1, + "cache_read": 0.2, + "cache_write": 0 } }, - "claude-opus-4-20250514": { - "id": "claude-opus-4-20250514", - "name": "Claude Opus 4", - "family": "claude-opus", - "attachment": true, + "GLM-5.1": { + "id": "GLM-5.1", + "name": "GLM-5.1", + "description": "General Language Model 5.1 — high-quality bilingual (EN/ZH) generation with strong coding and reasoning capabilities.", + "family": "glm", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2025-05-22", - "last_updated": "2025-05-22", + "knowledge": "2025-04", + "release_date": "2026-04-07", + "last_updated": "2026-06-01", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 200000, - "output": 32000 + "context": 202752, + "output": 131072 }, "cost": { - "input": 15, - "output": 75, - "cache_read": 1.5, - "cache_write": 18.75 + "input": 1, + "output": 3.2, + "cache_read": 0.1, + "cache_write": 0 } } } }, - "modelscope": { - "id": "modelscope", - "env": ["MODELSCOPE_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api-inference.modelscope.cn/v1", - "name": "ModelScope", - "doc": "https://modelscope.cn/docs/model-service/API-Inference/intro", + "minimax": { + "id": "minimax", + "env": ["MINIMAX_API_KEY"], + "npm": "@ai-sdk/anthropic", + "api": "https://api.minimax.io/anthropic/v1", + "name": "MiniMax (minimax.io)", + "doc": "https://platform.minimax.io/docs/guides/quickstart", "models": { - "Qwen/Qwen3-30B-A3B-Thinking-2507": { - "id": "Qwen/Qwen3-30B-A3B-Thinking-2507", - "name": "Qwen3 30B A3B Thinking 2507", - "family": "qwen", + "MiniMax-M2.1": { + "id": "MiniMax-M2.1", + "name": "MiniMax-M2.1", + "description": "Earlier MiniMax agent model for practical coding and productivity tasks", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "release_date": "2025-12-23", + "last_updated": "2025-12-23", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 32768 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2 } }, - "Qwen/Qwen3-30B-A3B-Instruct-2507": { - "id": "Qwen/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507", - "family": "qwen", + "MiniMax-M2.5-highspeed": { + "id": "MiniMax-M2.5-highspeed", + "name": "MiniMax-M2.5-highspeed", + "description": "High-speed MiniMax model for low-latency coding and agent workflows", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-30", - "last_updated": "2025-07-30", + "release_date": "2026-02-13", + "last_updated": "2026-02-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 16384 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "Qwen/Qwen3-235B-A22B-Instruct-2507": { - "id": "Qwen/Qwen3-235B-A22B-Instruct-2507", - "name": "Qwen3 235B A22B Instruct 2507", - "family": "qwen", + "MiniMax-M2.7-highspeed": { + "id": "MiniMax-M2.7-highspeed", + "name": "MiniMax-M2.7-highspeed", + "description": "Low-latency M2.7 variant for interactive coding plans and agent loops", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-04-28", - "last_updated": "2025-07-21", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 204800, "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.6, + "output": 2.4, + "cache_read": 0.06, + "cache_write": 0.375 } }, - "Qwen/Qwen3-Coder-30B-A3B-Instruct": { - "id": "Qwen/Qwen3-Coder-30B-A3B-Instruct", - "name": "Qwen3 Coder 30B A3B Instruct", - "family": "qwen", + "MiniMax-M2": { + "id": "MiniMax-M2", + "name": "MiniMax-M2", + "description": "Efficient open MiniMax model built for coding agents and tool-heavy workflows", + "family": "minimax", "attachment": false, - "reasoning": false, + "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-31", - "last_updated": "2025-07-31", + "release_date": "2025-10-27", + "last_updated": "2025-10-27", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 196608, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2 } }, - "Qwen/Qwen3-235B-A22B-Thinking-2507": { - "id": "Qwen/Qwen3-235B-A22B-Thinking-2507", - "name": "Qwen3-235B-A22B-Thinking-2507", - "family": "qwen", + "MiniMax-M2.5": { + "id": "MiniMax-M2.5", + "name": "MiniMax-M2.5", + "description": "Prior MiniMax coding model for agent workflows, office edits, and automation", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-25", - "last_updated": "2025-07-25", + "release_date": "2026-02-12", + "last_updated": "2026-02-12", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, + "context": 204800, "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2, + "cache_read": 0.03, + "cache_write": 0.375 } }, - "ZhipuAI/GLM-4.5": { - "id": "ZhipuAI/GLM-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, + "MiniMax-M3": { + "id": "MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", + "family": "minimax", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "release_date": "2026-06-01", + "last_updated": "2026-06-25", "modalities": { - "input": ["text"], + "input": ["text", "image", "video"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 1000000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "tiers": [ + { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12, + "tier": { + "type": "context", + "size": 512000 + } + } + ], + "context_over_200k": { + "input": 0.6, + "output": 2.4, + "cache_read": 0.12 + } } }, - "ZhipuAI/GLM-4.6": { - "id": "ZhipuAI/GLM-4.6", - "name": "GLM-4.6", - "family": "glm", + "MiniMax-M2.7": { + "id": "MiniMax-M2.7", + "name": "MiniMax-M2.7", + "description": "Open MiniMax flagship for coding agents, office automation, and complex environments", + "family": "minimax", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "knowledge": "2025-07", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", + "release_date": "2026-03-18", + "last_updated": "2026-03-18", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 202752, - "output": 98304 + "context": 204800, + "output": 131072 }, "cost": { - "input": 0, - "output": 0 + "input": 0.3, + "output": 1.2, + "cache_read": 0.06, + "cache_write": 0.375 } } } }, - "hpc-ai": { - "id": "hpc-ai", - "env": ["HPC_AI_API_KEY"], + "github-copilot": { + "id": "github-copilot", + "env": ["GITHUB_TOKEN"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.hpc-ai.com/inference/v1", - "name": "HPC-AI", - "doc": "https://www.hpc-ai.com/doc/docs/quickstart/", + "api": "https://api.githubcopilot.com", + "name": "GitHub Copilot", + "doc": "https://docs.github.com/en/copilot", "models": { - "zai-org/glm-5.1": { - "id": "zai-org/glm-5.1", - "name": "GLM 5.1", - "family": "glm", - "attachment": false, + "claude-sonnet-4.5": { + "id": "claude-sonnet-4.5", + "name": "Claude Sonnet 4.5 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "temperature": true, + "knowledge": "2025-07-31", + "release_date": "2025-09-29", + "last_updated": "2025-09-29", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "input": 168000, + "output": 32000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "claude-sonnet-4": { + "id": "claude-sonnet-4", + "name": "Claude Sonnet 4 (latest)", + "description": "Balanced Claude model for coding, analysis, agent workflows, and cost control", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "knowledge": "2025-03-31", + "release_date": "2025-05-22", + "last_updated": "2025-05-22", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 216000, + "input": 128000, + "output": 16000 }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "gemini-2.5-pro": { + "id": "gemini-2.5-pro", + "name": "Gemini 2.5 Pro", + "description": "Google's proven reasoning model for coding, math, and multimodal analysis", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 128, + "max": 32768 + } + ], + "tool_call": true, "structured_output": true, "temperature": true, - "release_date": "2026-04-08", - "last_updated": "2026-04-08", + "knowledge": "2025-01", + "release_date": "2025-06-17", + "last_updated": "2025-06-17", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 202000, - "output": 202000 + "context": 128000, + "input": 128000, + "output": 64000 }, "cost": { - "input": 0.66, - "output": 2, - "cache_read": 0.12 + "input": 1.25, + "output": 10, + "cache_read": 0.125 } }, - "minimax/minimax-m2.5": { - "id": "minimax/minimax-m2.5", - "name": "MiniMax M2.5", - "family": "minimax-m2.5", - "attachment": false, + "claude-haiku-4.5": { + "id": "claude-haiku-4.5", + "name": "Claude Haiku 4.5 (latest)", + "description": "Fast Claude lane for lightweight agents, office tasks, and responsive chat", + "family": "claude-haiku", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, - "structured_output": false, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-03-25", + "knowledge": "2025-02-28", + "release_date": "2025-10-15", + "last_updated": "2025-10-15", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1000000, - "output": 131072 + "context": 200000, + "input": 136000, + "output": 64000 }, "cost": { - "input": 0.14, - "output": 0.56, - "cache_read": 0.014 + "input": 1, + "output": 5, + "cache_read": 0.1, + "cache_write": 1.25 } }, - "moonshotai/kimi-k2.5": { - "id": "moonshotai/kimi-k2.5", - "name": "Kimi K2.5", - "family": "kimi", - "attachment": false, + "gemini-3.5-flash": { + "id": "gemini-3.5-flash", + "name": "Gemini 3.5 Flash", + "description": "Fast Gemini model balancing multimodal reasoning, tool use, and cost", + "family": "gemini-flash", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["minimal", "low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 256, + "max": 24000 + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-05-19", + "last_updated": "2026-05-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 200000, + "input": 128000, + "output": 64000 }, + "cost": { + "input": 1.5, + "output": 9, + "cache_read": 0.15, + "input_audio": 1.5 + } + }, + "kimi-k2.7-code": { + "id": "kimi-k2.7-code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", + "attachment": true, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2025-01-01", - "release_date": "2026-01-01", - "last_updated": "2026-03-25", + "knowledge": "2025-01", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text", "image", "video"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 262144 + "context": 256000, + "input": 224000, + "output": 32000 }, "cost": { - "input": 0.21, - "output": 1, - "cache_read": 0.03 + "input": 0.95, + "output": 4, + "cache_read": 0.19 } - } - } - }, - "gitlab": { - "id": "gitlab", - "env": ["GITLAB_TOKEN"], - "npm": "gitlab-ai-provider", - "name": "GitLab Duo", - "doc": "https://docs.gitlab.com/user/duo_agent_platform/", - "models": { - "duo-chat-gpt-5-4-nano": { - "id": "duo-chat-gpt-5-4-nano", - "name": "Agentic Chat (GPT-5.4 Nano)", + }, + "claude-sonnet-5": { + "id": "claude-sonnet-5", + "name": "Claude Sonnet 5", + "description": "Everyday Claude agent model for coding, planning, browsing, and general work", + "family": "claude-sonnet", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "temperature": false, + "knowledge": "2026-01-31", + "release_date": "2026-06-30", + "last_updated": "2026-06-30", + "modalities": { + "input": ["text", "image", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "output": 128000 + }, + "cost": { + "input": 2, + "output": 10, + "cache_read": 0.2, + "cache_write": 2.5 + } + }, + "gpt-5.4-nano": { + "id": "gpt-5.4-nano", + "name": "GPT-5.4 nano", + "description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation", "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": false, @@ -115339,76 +170275,115 @@ "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.2, + "output": 1.25, + "cache_read": 0.02 } }, - "duo-chat-gpt-5-mini": { - "id": "duo-chat-gpt-5-mini", - "name": "Agentic Chat (GPT-5 Mini)", - "family": "gpt-mini", + "claude-opus-4.7": { + "id": "claude-opus-4.7", + "name": "Claude Opus 4.7", + "description": "Stronger Opus tier for advanced software work and high-stakes reasoning", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2024-05-30", - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "knowledge": "2026-01-31", + "release_date": "2026-04-16", + "last_updated": "2026-04-16", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, - "output": 128000 + "context": 200000, + "input": 168000, + "output": 32000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 0, - "output": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "duo-chat-sonnet-4-6": { - "id": "duo-chat-sonnet-4-6", - "name": "Agentic Chat (Claude Sonnet 4.6)", - "family": "claude-sonnet", - "attachment": true, + "mai-code-1-flash-picker": { + "id": "mai-code-1-flash-picker", + "name": "MAI-Code-1-Flash", + "description": "Microsoft coding model built for fast, efficient assistance in everyday developer workflows", + "family": "mai", + "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "structured_output": true, "temperature": true, - "knowledge": "2025-08-31", - "release_date": "2026-02-17", - "last_updated": "2026-02-17", + "knowledge": "2025-12", + "release_date": "2026-06-02", + "last_updated": "2026-06-08", "modalities": { - "input": ["text", "image", "pdf"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 256000, + "input": 128000, + "output": 128000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "duo-chat-gpt-5-2": { - "id": "duo-chat-gpt-5-2", - "name": "Agentic Chat (GPT-5.2)", + "gpt-5.2": { + "id": "gpt-5.2", + "name": "GPT-5.2", + "description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work", "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-01-23", - "last_updated": "2026-01-23", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -115420,24 +170395,32 @@ "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "duo-chat-gpt-5-codex": { - "id": "duo-chat-gpt-5-codex", - "name": "Agentic Chat (GPT-5 Codex)", + "gpt-5.3-codex": { + "id": "gpt-5.3-codex", + "name": "GPT-5.3 Codex", + "description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work", "family": "gpt-codex", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "knowledge": "2025-08-31", + "release_date": "2026-02-05", + "last_updated": "2026-02-05", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, @@ -115447,75 +170430,131 @@ "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "duo-chat-gpt-5-1": { - "id": "duo-chat-gpt-5-1", - "name": "Agentic Chat (GPT-5.1)", - "family": "gpt", + "gpt-5.6-luna": { + "id": "gpt-5.6-luna", + "name": "GPT-5.6 Luna", + "description": "Cost-efficient GPT-5.6 model for fast, high-volume workloads", + "family": "gpt-nano", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2024-09-30", - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text", "image"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 1, + "output": 6, + "cache_read": 0.1, + "tiers": [ + { + "input": 2, + "output": 9, + "cache_read": 0.2, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 2, + "output": 9, + "cache_read": 0.2 + } } }, - "duo-chat-gpt-5-2-codex": { - "id": "duo-chat-gpt-5-2-codex", - "name": "Agentic Chat (GPT-5.2 Codex)", - "family": "gpt-codex", + "gpt-5.6-terra": { + "id": "gpt-5.6-terra", + "name": "GPT-5.6 Terra", + "description": "Balanced GPT-5.6 model for capable, cost-efficient everyday work", + "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-01-22", - "last_updated": "2026-01-22", + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "duo-chat-sonnet-4-5": { - "id": "duo-chat-sonnet-4-5", - "name": "Agentic Chat (Claude Sonnet 4.5)", - "family": "claude-sonnet", + "claude-opus-4.8": { + "id": "claude-opus-4.8", + "name": "Claude Opus 4.8", + "description": "Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "temperature": true, - "knowledge": "2025-07-31", - "release_date": "2026-01-08", - "last_updated": "2026-01-08", + "temperature": false, + "knowledge": "2026-01", + "release_date": "2026-05-28", + "last_updated": "2026-05-28", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -115523,53 +170562,89 @@ "open_weights": false, "limit": { "context": 200000, + "input": 168000, "output": 64000 }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } + }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "duo-chat-gpt-5-4": { - "id": "duo-chat-gpt-5-4", - "name": "Agentic Chat (GPT-5.4)", - "family": "gpt", + "claude-fable-5": { + "id": "claude-fable-5", + "name": "Claude Fable 5", + "description": "Claude model for creative writing, analysis, and controlled agent workflows", + "family": "claude-fable", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "structured_output": true, "temperature": false, - "knowledge": "2025-08-31", - "release_date": "2026-03-05", - "last_updated": "2026-03-05", + "knowledge": "2026-01-31", + "release_date": "2026-06-09", + "last_updated": "2026-06-09", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1050000, - "input": 922000, + "context": 1000000, "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 10, + "output": 50, + "cache_read": 1, + "cache_write": 12.5 } }, - "duo-chat-haiku-4-5": { - "id": "duo-chat-haiku-4-5", - "name": "Agentic Chat (Claude Haiku 4.5)", - "family": "claude-haiku", + "claude-opus-4.5": { + "id": "claude-opus-4.5", + "name": "Claude Opus 4.5 (latest)", + "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents", + "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-02-28", - "release_date": "2026-01-08", - "last_updated": "2026-01-08", + "knowledge": "2025-05", + "release_date": "2025-11-24", + "last_updated": "2025-11-24", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -115577,48 +170652,80 @@ "open_weights": false, "limit": { "context": 200000, - "output": 64000 + "input": 168000, + "output": 32000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } }, - "duo-chat-gpt-5-3-codex": { - "id": "duo-chat-gpt-5-3-codex", - "name": "Agentic Chat (GPT-5.3 Codex)", - "family": "gpt-codex", + "gpt-5.4": { + "id": "gpt-5.4", + "name": "GPT-5.4", + "description": "Agent-ready GPT for coding and computer-use workflows at a lower cost", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, "knowledge": "2025-08-31", - "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "release_date": "2026-03-05", + "last_updated": "2026-03-05", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 400000, - "input": 272000, + "context": 1050000, + "input": 922000, "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 2.5, + "output": 15, + "cache_read": 0.25, + "tiers": [ + { + "input": 5, + "output": 22.5, + "cache_read": 0.5, + "tier": { + "type": "context", + "size": 272000 + } + } + ], + "context_over_200k": { + "input": 5, + "output": 22.5, + "cache_read": 0.5 + } } }, - "duo-chat-gpt-5-4-mini": { - "id": "duo-chat-gpt-5-4-mini", - "name": "Agentic Chat (GPT-5.4 Mini)", + "gpt-5.4-mini": { + "id": "gpt-5.4-mini", + "name": "GPT-5.4 mini", + "description": "Strong small GPT for coding subagents, quick tool use, and high-volume work", "family": "gpt-mini", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, "structured_output": true, "temperature": false, @@ -115636,48 +170743,119 @@ "output": 128000 }, "cost": { - "input": 0, - "output": 0 + "input": 0.75, + "output": 4.5, + "cache_read": 0.075 } }, - "duo-chat-opus-4-7": { - "id": "duo-chat-opus-4-7", - "name": "Agentic Chat (Claude Opus 4.7)", - "family": "claude-opus", + "gpt-4.1": { + "id": "gpt-4.1", + "name": "GPT-4.1", + "description": "Long-lived GPT workhorse for coding, instruction following, and production apps", + "family": "gpt", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, - "temperature": false, - "knowledge": "2026-01-31", - "release_date": "2026-04-16", - "last_updated": "2026-04-16", + "structured_output": true, + "temperature": true, + "knowledge": "2024-04", + "release_date": "2025-04-14", + "last_updated": "2025-04-14", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, + "limit": { + "context": 128000, + "input": 128000, + "output": 16384 + }, + "cost": { + "input": 2, + "output": 8, + "cache_read": 0.5 + } + }, + "gemini-3.1-pro-preview": { + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro Preview", + "description": "Reasoning-first Gemini preview for agentic coding and complex problem solving", + "family": "gemini-pro", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 256, + "max": 32000 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2026-02-19", + "last_updated": "2026-02-19", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, "limit": { "context": 1000000, + "input": 936000, "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 2, + "output": 12, + "cache_read": 0.2, + "tiers": [ + { + "input": 4, + "output": 18, + "cache_read": 0.4, + "tier": { + "type": "context", + "size": 200000 + } + } + ], + "context_over_200k": { + "input": 4, + "output": 18, + "cache_read": 0.4 + } } }, - "duo-chat-opus-4-5": { - "id": "duo-chat-opus-4-5", - "name": "Agentic Chat (Claude Opus 4.5)", - "family": "claude-opus", + "claude-sonnet-4.6": { + "id": "claude-sonnet-4.6", + "name": "Claude Sonnet 4.6", + "description": "Claude workhorse for coding agents, careful analysis, and production cost control", + "family": "claude-sonnet", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + }, + { + "type": "budget_tokens", + "min": 1024, + "max": 32000 + } + ], "tool_call": true, "temperature": true, - "knowledge": "2025-03-31", - "release_date": "2026-01-08", - "last_updated": "2026-01-08", + "knowledge": "2025-08-31", + "release_date": "2026-02-17", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] @@ -115685,266 +170863,394 @@ "open_weights": false, "limit": { "context": 200000, + "input": 168000, + "output": 32000 + }, + "cost": { + "input": 3, + "output": 15, + "cache_read": 0.3, + "cache_write": 3.75 + } + }, + "gpt-5-mini": { + "id": "gpt-5-mini", + "name": "GPT-5 Mini", + "description": "Small GPT-5 for responsive agents, coding help, and everyday automation", + "family": "gpt-mini", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "structured_output": true, + "temperature": false, + "knowledge": "2024-05-30", + "release_date": "2025-08-07", + "last_updated": "2025-08-07", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 264000, + "input": 128000, "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 0.25, + "output": 2, + "cache_read": 0.025 } }, - "duo-chat-opus-4-6": { - "id": "duo-chat-opus-4-6", - "name": "Agentic Chat (Claude Opus 4.6)", + "gemini-3-flash-preview": { + "id": "gemini-3-flash-preview", + "name": "Gemini 3 Flash Preview", + "description": "New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs", + "family": "gemini-flash", + "attachment": true, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + }, + { + "type": "budget_tokens", + "min": 256, + "max": 32000 + } + ], + "tool_call": true, + "structured_output": true, + "temperature": true, + "knowledge": "2025-01", + "release_date": "2025-12-17", + "last_updated": "2025-12-17", + "modalities": { + "input": ["text", "image", "video", "audio", "pdf"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 128000, + "output": 64000 + }, + "cost": { + "input": 0.5, + "output": 3, + "cache_read": 0.05, + "input_audio": 1 + } + }, + "claude-opus-4.6": { + "id": "claude-opus-4.6", + "name": "Claude Opus 4.6", + "description": "High-end Claude for difficult coding, planning, and slower expert reasoning", "family": "claude-opus", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high", "max"] + } + ], "tool_call": true, "temperature": true, "knowledge": "2025-05-31", "release_date": "2026-02-05", - "last_updated": "2026-02-05", + "last_updated": "2026-03-13", "modalities": { "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, - "output": 64000 + "context": 200000, + "input": 168000, + "output": 32000 + }, + "experimental": { + "modes": { + "fast": { + "cost": { + "input": 30, + "output": 150, + "cache_read": 3, + "cache_write": 37.5 + }, + "provider": { + "body": { + "speed": "fast" + }, + "headers": { + "anthropic-beta": "fast-mode-2026-02-01" + } + } + } + } }, "cost": { - "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "input": 5, + "output": 25, + "cache_read": 0.5, + "cache_write": 6.25 } - } - } - }, - "xiaomi": { - "id": "xiaomi", - "env": ["XIAOMI_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.xiaomimimo.com/v1", - "name": "Xiaomi", - "doc": "https://platform.xiaomimimo.com/#/docs", - "models": { - "mimo-v2.5-pro": { - "id": "mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", - "family": "mimo", - "attachment": false, + }, + "gpt-5.6-sol": { + "id": "gpt-5.6-sol", + "name": "GPT-5.6 Sol", + "description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows", + "family": "gpt", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh", "max"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "structured_output": true, + "temperature": false, + "knowledge": "2026-02-16", + "release_date": "2026-07-09", + "last_updated": "2026-07-09", "modalities": { - "input": ["text"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, + "input": 5, + "output": 30, + "cache_read": 0.5, "tiers": [ { - "input": 2, - "output": 6, - "cache_read": 0.4, + "input": 10, + "output": 45, + "cache_read": 1, "tier": { "type": "context", - "size": 256000 + "size": 272000 } } ], "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 + "input": 10, + "output": 45, + "cache_read": 1 } } }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo-V2-Omni", - "family": "mimo", + "gpt-5.2-codex": { + "id": "gpt-5.2-codex", + "name": "GPT-5.2 Codex", + "description": "Code-specialist GPT for repository edits, reviews, and long-running software agents", + "family": "gpt-codex", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "structured_output": true, + "temperature": false, + "knowledge": "2025-08-31", + "release_date": "2025-12-11", + "last_updated": "2025-12-11", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], + "input": ["text", "image", "pdf"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 262144, - "output": 131072 + "context": 400000, + "input": 272000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08 + "input": 1.75, + "output": 14, + "cache_read": 0.175 } }, - "mimo-v2.5": { - "id": "mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", + "gpt-5.5": { + "id": "gpt-5.5", + "name": "GPT-5.5", + "description": "Default frontier GPT for coding, computer use, research, and knowledge work", + "family": "gpt", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high", "xhigh"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", + "structured_output": true, + "temperature": false, + "knowledge": "2025-12-01", + "release_date": "2026-04-23", + "last_updated": "2026-04-23", "modalities": { - "input": ["text", "image", "audio", "video"], + "input": ["text", "image", "pdf"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 1048576, - "output": 131072 + "context": 1050000, + "input": 922000, + "output": 128000 }, "cost": { - "input": 0.4, - "output": 2, - "cache_read": 0.08, + "input": 5, + "output": 30, + "cache_read": 0.5, "tiers": [ { - "input": 0.8, - "output": 4, - "cache_read": 0.16, + "input": 10, + "output": 45, + "cache_read": 1, "tier": { "type": "context", - "size": 256000 + "size": 272000 } } ], "context_over_200k": { - "input": 0.8, - "output": 4, - "cache_read": 0.16 + "input": 10, + "output": 45, + "cache_read": 1 } } - }, - "mimo-v2-pro": { - "id": "mimo-v2-pro", - "name": "MiMo-V2-Pro", - "family": "mimo", - "attachment": false, + } + } + }, + "clarifai": { + "id": "clarifai", + "env": ["CLARIFAI_PAT"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.clarifai.com/v2/ext/openai/v1", + "name": "Clarifai", + "doc": "https://docs.clarifai.com/compute/inference/", + "models": { + "moonshotai/chat-completion/models/Kimi-K2_6": { + "id": "moonshotai/chat-completion/models/Kimi-K2_6", + "name": "Kimi K2.6", + "description": "Multimodal Kimi workhorse for agent loops, coding tasks, and visual context", + "family": "kimi-k2", + "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "interleaved": { "field": "reasoning_content" }, + "structured_output": true, "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "knowledge": "2025-01", + "release_date": "2026-04-21", + "last_updated": "2026-04-21", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": true, + "limit": { + "context": 262144, + "output": 262144 + }, + "cost": { + "input": 0.95, + "output": 4 + } + }, + "minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput": { + "id": "minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput", + "name": "MiniMax-M2.5 High Throughput", + "description": "MiniMax model for chat, coding, office work, and agentic tasks", + "family": "minimax", + "attachment": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, + "temperature": true, + "release_date": "2026-02-12", + "last_updated": "2026-02-25", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 1048576, + "context": 204800, "output": 131072 }, "cost": { - "input": 1, - "output": 3, - "cache_read": 0.2, - "tiers": [ - { - "input": 2, - "output": 6, - "cache_read": 0.4, - "tier": { - "type": "context", - "size": 256000 - } - } - ], - "context_over_200k": { - "input": 2, - "output": 6, - "cache_read": 0.4 - } + "input": 0.3, + "output": 1.2 } }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "MiMo-V2-Flash", - "family": "mimo", + "openai/chat-completion/models/gpt-oss-120b-high-throughput": { + "id": "openai/chat-completion/models/gpt-oss-120b-high-throughput", + "name": "GPT OSS 120B High Throughput", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "release_date": "2025-08-05", + "last_updated": "2026-02-25", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 131072, + "output": 16384 }, "cost": { - "input": 0.1, - "output": 0.3, - "cache_read": 0.01 + "input": 0.09, + "output": 0.36 } - } - } - }, - "clarifai": { - "id": "clarifai", - "env": ["CLARIFAI_PAT"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://api.clarifai.com/v2/ext/openai/v1", - "name": "Clarifai", - "doc": "https://docs.clarifai.com/compute/inference/", - "models": { - "arcee_ai/AFM/models/trinity-mini": { - "id": "arcee_ai/AFM/models/trinity-mini", - "name": "Trinity Mini", - "family": "trinity-mini", + }, + "openai/chat-completion/models/gpt-oss-20b": { + "id": "openai/chat-completion/models/gpt-oss-20b", + "name": "GPT OSS 20B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "knowledge": "2024-10", - "release_date": "2025-12", - "last_updated": "2026-02-25", + "release_date": "2025-08-05", + "last_updated": "2025-12-12", "modalities": { "input": ["text"], "output": ["text"] @@ -115952,19 +171258,21 @@ "open_weights": true, "limit": { "context": 131072, - "output": 131072 + "output": 16384 }, "cost": { "input": 0.045, - "output": 0.15 + "output": 0.18 } }, "mistralai/completion/models/Ministral-3-14B-Reasoning-2512": { "id": "mistralai/completion/models/Ministral-3-14B-Reasoning-2512", "name": "Ministral 3 14B Reasoning 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", "family": "ministral", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "knowledge": "2025-12", @@ -115987,9 +171295,11 @@ "mistralai/completion/models/Ministral-3-3B-Reasoning-2512": { "id": "mistralai/completion/models/Ministral-3-3B-Reasoning-2512", "name": "Ministral 3 3B Reasoning 2512", + "description": "Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads", "family": "ministral", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, "release_date": "2025-12", @@ -116011,6 +171321,7 @@ "deepseek-ai/deepseek-ocr/models/DeepSeek-OCR": { "id": "deepseek-ai/deepseek-ocr/models/DeepSeek-OCR", "name": "DeepSeek OCR", + "description": "OCR model for extracting structured text from documents and screenshots", "family": "deepseek", "attachment": true, "reasoning": false, @@ -116032,39 +171343,18 @@ "output": 0.7 } }, - "openai/chat-completion/models/gpt-oss-20b": { - "id": "openai/chat-completion/models/gpt-oss-20b", - "name": "GPT OSS 20B", - "family": "gpt-oss", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-12-12", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 16384 - }, - "cost": { - "input": 0.045, - "output": 0.18 - } - }, - "openai/chat-completion/models/gpt-oss-120b-high-throughput": { - "id": "openai/chat-completion/models/gpt-oss-120b-high-throughput", - "name": "GPT OSS 120B High Throughput", - "family": "gpt-oss", + "qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507": { + "id": "qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507", + "name": "Qwen3 30B A3B Thinking 2507", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2025-08-05", + "release_date": "2025-07-31", "last_updated": "2026-02-25", "modalities": { "input": ["text"], @@ -116072,23 +171362,25 @@ }, "open_weights": true, "limit": { - "context": 131072, - "output": 16384 + "context": 262144, + "output": 131072 }, "cost": { - "input": 0.09, - "output": 0.36 + "input": 0.36, + "output": 1.3 } }, - "minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput": { - "id": "minimaxai/chat-completion/models/MiniMax-M2_5-high-throughput", - "name": "MiniMax-M2.5 High Throughput", - "family": "minimax", + "qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507": { + "id": "qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507", + "name": "Qwen3 30B A3B Instruct 2507", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", + "release_date": "2025-07-30", "last_updated": "2026-02-25", "modalities": { "input": ["text"], @@ -116096,17 +171388,18 @@ }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 262144 }, "cost": { "input": 0.3, - "output": 1.2 + "output": 0.5 } }, "qwen/qwenCoder/models/Qwen3-Coder-30B-A3B-Instruct": { "id": "qwen/qwenCoder/models/Qwen3-Coder-30B-A3B-Instruct", "name": "Qwen3 Coder 30B A3B Instruct", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", "family": "qwen", "attachment": false, "reasoning": false, @@ -116129,16 +171422,18 @@ "output": 0.74812 } }, - "qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507": { - "id": "qwen/qwenLM/models/Qwen3-30B-A3B-Thinking-2507", - "name": "Qwen3 30B A3B Thinking 2507", - "family": "qwen", + "arcee_ai/AFM/models/trinity-mini": { + "id": "arcee_ai/AFM/models/trinity-mini", + "name": "Trinity Mini", + "description": "Efficient model for low-latency assistance, extraction, and routine automation", + "family": "trinity-mini", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "structured_output": true, "temperature": true, - "release_date": "2025-07-31", + "knowledge": "2024-10", + "release_date": "2025-12", "last_updated": "2026-02-25", "modalities": { "input": ["text"], @@ -116146,78 +171441,375 @@ }, "open_weights": true, "limit": { - "context": 262144, + "context": 131072, "output": 131072 }, "cost": { - "input": 0.36, - "output": 1.3 + "input": 0.045, + "output": 0.15 } }, - "qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507": { - "id": "qwen/qwenLM/models/Qwen3-30B-A3B-Instruct-2507", - "name": "Qwen3 30B A3B Instruct 2507", - "family": "qwen", - "attachment": false, + "clarifai/main/models/mm-poly-8b": { + "id": "clarifai/main/models/mm-poly-8b", + "name": "MM Poly 8B", + "description": "Multimodal model for analyzing text, images, documents, and rich media", + "family": "mm-poly", + "attachment": true, "reasoning": false, + "tool_call": false, + "temperature": true, + "release_date": "2025-06", + "last_updated": "2026-02-25", + "modalities": { + "input": ["text", "image", "video"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 32768, + "output": 4096 + }, + "cost": { + "input": 0.658, + "output": 1.11 + } + } + } + }, + "the-grid-ai": { + "id": "the-grid-ai", + "env": ["THEGRIDAI_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.thegrid.ai/v1", + "name": "The Grid AI", + "doc": "https://thegrid.ai/docs", + "models": { + "agent-prime": { + "id": "agent-prime", + "name": "Agent Prime", + "description": "Reliable models for dependable agentic applications, multi-step tool use, and reasoning workflows. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "structured_output": true, "temperature": true, - "release_date": "2025-07-30", - "last_updated": "2026-02-25", + "release_date": "2026-05-04", + "last_updated": "2026-07-06", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 262144 + "context": 196608, + "input": 120000, + "output": 30000 }, - "cost": { - "input": 0.3, - "output": 0.5 + "status": "beta" + }, + "agent-max": { + "id": "agent-max", + "name": "Agent Max", + "description": "Frontier models for autonomous research, deep multi-step tool chains, and complex long-horizon tasks. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-04", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 128000 + }, + "status": "beta" + }, + "text-standard": { + "id": "text-standard", + "name": "Text Standard", + "description": "Price-optimized models with low-latency, high-throughput and shorter maximum outputs. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 120000, + "output": 16000 } }, - "clarifai/main/models/mm-poly-8b": { - "id": "clarifai/main/models/mm-poly-8b", - "name": "MM Poly 8B", - "family": "mm-poly", - "attachment": true, - "reasoning": false, - "tool_call": false, + "code-prime": { + "id": "code-prime", + "name": "Code Prime", + "description": "Reliable models for everyday software tasks, code completion, review, and standard debugging. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-04", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 196608, + "input": 120000, + "output": 30000 + }, + "status": "beta" + }, + "text-prime": { + "id": "text-prime", + "name": "Text Prime", + "description": "Reliable models for everyday text generation, editing, and analysis across diverse workflows. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-02-26", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 196608, + "input": 120000, + "output": 30000 + } + }, + "code-max": { + "id": "code-max", + "name": "Code Max", + "description": "Frontier models for complex research, architectural decisions, debugging, and multi-file development. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-04", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 128000 + }, + "status": "beta" + }, + "agent-standard": { + "id": "agent-standard", + "name": "Agent Standard", + "description": "Price-optimized models for fast tool calls, simple agent loops, high-throughput automation, and orchestration. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-05-04", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 128000, + "input": 120000, + "output": 16000 + }, + "status": "beta" + }, + "text-max": { + "id": "text-max", + "name": "Text Max", + "description": "Frontier models for deep reasoning, long context, and complex workflows. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + }, + { + "type": "effort", + "values": ["low", "medium", "high", "xhigh", "max"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, + "temperature": true, + "release_date": "2026-03-24", + "last_updated": "2026-07-06", + "modalities": { + "input": ["text", "image"], + "output": ["text"] + }, + "open_weights": false, + "limit": { + "context": 1000000, + "input": 922000, + "output": 128000 + } + }, + "code-standard": { + "id": "code-standard", + "name": "Code Standard", + "description": "Price-optimized models for rapid autocomplete, linting, high-frequency suggestions, and batch edits. Any model that meets the contract spec can serve your request.", + "attachment": false, + "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], + "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-06", - "last_updated": "2026-02-25", + "release_date": "2026-05-04", + "last_updated": "2026-07-06", "modalities": { - "input": ["text", "image", "video"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 32768, - "output": 4096 + "context": 128000, + "input": 120000, + "output": 16000 }, - "cost": { - "input": 0.658, - "output": 1.11 - } - }, - "moonshotai/chat-completion/models/Kimi-K2_6": { - "id": "moonshotai/chat-completion/models/Kimi-K2_6", - "name": "Kimi K2.6", - "family": "kimi-k2.6", + "status": "beta" + } + } + }, + "synthetic": { + "id": "synthetic", + "env": ["SYNTHETIC_API_KEY"], + "npm": "@ai-sdk/openai-compatible", + "api": "https://api.synthetic.new/openai/v1", + "name": "Synthetic", + "doc": "https://synthetic.new/pricing", + "models": { + "hf:moonshotai/Kimi-K2.7-Code": { + "id": "hf:moonshotai/Kimi-K2.7-Code", + "name": "Kimi K2.7 Code", + "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking", + "family": "kimi-k2", "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "structured_output": true, - "temperature": true, + "temperature": false, "knowledge": "2025-01", - "release_date": "2026-04-21", - "last_updated": "2026-04-21", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { "input": ["text", "image"], "output": ["text"] @@ -116225,33 +171817,34 @@ "open_weights": true, "limit": { "context": 262144, - "output": 262144 + "output": 65536 }, "cost": { "input": 0.95, - "output": 4 + "output": 4, + "cache_read": 0.95 } - } - } - }, - "minimax-cn": { - "id": "minimax-cn", - "env": ["MINIMAX_API_KEY"], - "npm": "@ai-sdk/anthropic", - "api": "https://api.minimaxi.com/anthropic/v1", - "name": "MiniMax (minimaxi.com)", - "doc": "https://platform.minimaxi.com/docs/guides/quickstart", - "models": { - "MiniMax-M2": { - "id": "MiniMax-M2", - "name": "MiniMax-M2", - "family": "minimax", + }, + "hf:zai-org/GLM-4.7-Flash": { + "id": "hf:zai-org/GLM-4.7-Flash", + "name": "GLM-4.7-Flash", + "description": "Budget GLM lane for fast coding help, routing, and everyday automation", + "family": "glm-flash", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2025-10-27", - "last_updated": "2025-10-27", + "knowledge": "2025-04", + "release_date": "2026-01-19", + "last_updated": "2026-01-19", "modalities": { "input": ["text"], "output": ["text"] @@ -116259,209 +171852,263 @@ "open_weights": true, "limit": { "context": 196608, - "output": 128000 + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.1, + "output": 0.5, + "cache_read": 0.1 } }, - "MiniMax-M2.5": { - "id": "MiniMax-M2.5", - "name": "MiniMax-M2.5", - "family": "minimax", + "hf:zai-org/GLM-5.2": { + "id": "hf:zai-org/GLM-5.2", + "name": "GLM-5.2", + "description": "Open flagship GLM for long-horizon coding agents and million-token context work", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "high", "xhigh"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-02-12", - "last_updated": "2026-02-12", + "release_date": "2026-06-13", + "last_updated": "2026-06-13", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 524288, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2, - "cache_read": 0.03, - "cache_write": 0.375 + "input": 1.4, + "output": 4.4, + "cache_read": 1.4 } }, - "MiniMax-M2.7": { - "id": "MiniMax-M2.7", - "name": "MiniMax-M2.7", + "hf:MiniMaxAI/MiniMax-M3": { + "id": "hf:MiniMaxAI/MiniMax-M3", + "name": "MiniMax-M3", + "description": "MiniMax multimodal model for long-context coding, perception, and agent planning", "family": "minimax", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-06-12", + "last_updated": "2026-06-12", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 524288, + "output": 65536 }, "cost": { - "input": 0.3, + "input": 0.6, "output": 1.2, - "cache_read": 0.06, - "cache_write": 0.375 + "cache_read": 0.6 } }, - "MiniMax-M2.7-highspeed": { - "id": "MiniMax-M2.7-highspeed", - "name": "MiniMax-M2.7-highspeed", - "family": "minimax", + "hf:openai/gpt-oss-120b": { + "id": "hf:openai/gpt-oss-120b", + "name": "GPT OSS 120B", + "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads", + "family": "gpt-oss", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, "temperature": true, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2025-08-05", + "last_updated": "2025-08-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 131072, + "output": 32768 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.1, + "output": 0.1, + "cache_read": 0.1 } }, - "MiniMax-M2.1": { - "id": "MiniMax-M2.1", - "name": "MiniMax-M2.1", - "family": "minimax", - "attachment": false, + "hf:Qwen/Qwen3.6-27B": { + "id": "hf:Qwen/Qwen3.6-27B", + "name": "Qwen3.6 27B", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", + "family": "qwen", + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["none", "low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, + "structured_output": true, "temperature": true, - "release_date": "2025-12-23", - "last_updated": "2025-12-23", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0.45, + "output": 3.6, + "cache_read": 0.45 } }, - "MiniMax-M2.5-highspeed": { - "id": "MiniMax-M2.5-highspeed", - "name": "MiniMax-M2.5-highspeed", - "family": "minimax", + "hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4": { + "id": "hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4", + "name": "Nemotron 3 Super 120B A12B", + "description": "Nemotron middle tier for collaborative agents and high-volume reasoning workloads", + "family": "nemotron", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "effort", + "values": ["low", "medium", "high"] + } + ], "tool_call": true, + "interleaved": { + "field": "reasoning_content" + }, "temperature": true, - "release_date": "2026-02-13", - "last_updated": "2026-02-13", + "release_date": "2026-03-11", + "last_updated": "2026-03-11", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 204800, - "output": 131072 + "context": 262144, + "output": 65536 }, "cost": { - "input": 0.6, - "output": 2.4, - "cache_read": 0.06, - "cache_write": 0.375 + "input": 0.3, + "output": 1, + "cache_read": 0.3 } } } }, - "regolo-ai": { - "id": "regolo-ai", - "env": ["REGOLO_API_KEY"], + "iflowcn": { + "id": "iflowcn", + "env": ["IFLOW_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.regolo.ai/v1", - "name": "Regolo AI", - "doc": "https://docs.regolo.ai/", + "api": "https://apis.iflow.cn/v1", + "name": "iFlow", + "doc": "https://platform.iflow.cn/en/docs", "models": { - "mistral-small3.2": { - "id": "mistral-small3.2", - "name": "Mistral Small 3.2", - "family": "mistral-small", + "qwen3-coder-plus": { + "id": "qwen3-coder-plus", + "name": "Qwen3-Coder-Plus", + "description": "Qwen coding model for software agents, repository edits, and code reasoning", + "family": "qwen", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-01-31", - "last_updated": "2025-01-31", + "knowledge": "2025-04", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 120000, - "output": 120000 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.5, - "output": 2.2 + "input": 0, + "output": 0 } }, - "qwen3-embedding-8b": { - "id": "qwen3-embedding-8b", - "name": "Qwen3-Embedding-8B", - "family": "qwen", + "deepseek-v3": { + "id": "deepseek-v3", + "name": "DeepSeek-V3", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-26", + "last_updated": "2024-12-26", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 8192 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.1, - "output": 0.1 + "input": 0, + "output": 0 } }, - "llama-3.3-70b-instruct": { - "id": "llama-3.3-70b-instruct", - "name": "Llama 3.3 70B Instruct", - "family": "llama", + "kimi-k2": { + "id": "kimi-k2", + "name": "Kimi-K2", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-04-28", - "last_updated": "2025-04-28", + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] @@ -116469,167 +172116,185 @@ "open_weights": false, "limit": { "context": 128000, - "output": 16384 + "output": 64000 }, "cost": { - "input": 0.6, - "output": 2.7 + "input": 0, + "output": 0 } }, - "qwen3-reranker-4b": { - "id": "qwen3-reranker-4b", - "name": "Qwen3-Reranker-4B", + "qwen3-32b": { + "id": "qwen3-32b", + "name": "Qwen3-32B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, "reasoning": false, - "tool_call": false, - "temperature": false, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "tool_call": true, + "temperature": true, + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 32768, - "output": 8192 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.12, - "output": 0.12 + "input": 0, + "output": 0 } }, - "mistral-small-4-119b": { - "id": "mistral-small-4-119b", - "name": "Mistral Small 4 119B", - "family": "mistral-small", - "attachment": true, - "reasoning": true, + "qwen3-max-preview": { + "id": "qwen3-max-preview", + "name": "Qwen3-Max-Preview", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", + "family": "qwen", + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-03-15", - "last_updated": "2026-03-15", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { "context": 256000, - "output": 16384 + "output": 32000 }, "cost": { - "input": 0.75, - "output": 3 + "input": 0, + "output": 0 } }, - "qwen3.5-122b": { - "id": "qwen3.5-122b", - "name": "Qwen3.5-122B", + "qwen3-max": { + "id": "qwen3-max", + "name": "Qwen3-Max", + "description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows", "family": "qwen", - "attachment": true, - "reasoning": true, + "attachment": false, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { - "input": ["text", "image"], + "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 16384 + "context": 256000, + "output": 32000 }, "cost": { - "input": 0.9, - "output": 3.6 + "input": 0, + "output": 0 } }, - "qwen-image": { - "id": "qwen-image", - "name": "Qwen-Image", + "qwen3-235b": { + "id": "qwen3-235b", + "name": "Qwen3-235B-A22B", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", "family": "qwen", "attachment": false, - "reasoning": false, - "tool_call": false, + "reasoning": true, + "reasoning_options": [], + "tool_call": true, "temperature": true, - "release_date": "2026-03-01", - "last_updated": "2026-03-01", + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2024-12-01", "modalities": { "input": ["text"], - "output": ["image"] + "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 8192, - "output": 4096 + "context": 128000, + "output": 32000 }, "cost": { - "input": 0.5, - "output": 2 + "input": 0, + "output": 0 } }, - "qwen3-coder-next": { - "id": "qwen3-coder-next", - "name": "Qwen3-Coder-Next", - "family": "qwen", + "glm-4.6": { + "id": "glm-4.6", + "name": "GLM-4.6", + "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering", + "family": "glm", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-03-01", - "last_updated": "2026-03-01", + "knowledge": "2024-10", + "release_date": "2024-12-01", + "last_updated": "2025-11-13", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": true, + "open_weights": false, "limit": { - "context": 262144, - "output": 16384 + "context": 200000, + "output": 128000 }, "cost": { - "input": 0.3, - "output": 1.2 + "input": 0, + "output": 0 } }, - "minimax-m2.5": { - "id": "minimax-m2.5", - "name": "MiniMax 2.5", - "family": "minimax", + "qwen3-235b-a22b-thinking-2507": { + "id": "qwen3-235b-a22b-thinking-2507", + "name": "Qwen3-235B-A22B-Thinking", + "description": "Qwen reasoning model for deliberate problem solving, math, and coding", + "family": "qwen", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-03-10", - "last_updated": "2026-03-10", + "knowledge": "2025-04", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 190000, + "context": 256000, "output": 64000 }, "cost": { - "input": 0.8, - "output": 3.5 + "input": 0, + "output": 0 } }, - "gpt-oss-20b": { - "id": "gpt-oss-20b", - "name": "GPT-OSS-20B", - "family": "gpt-oss", + "deepseek-r1": { + "id": "deepseek-r1", + "name": "DeepSeek-R1", + "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools", + "family": "deepseek-thinking", "attachment": false, "reasoning": true, + "reasoning_options": [], "tool_call": true, "temperature": true, - "release_date": "2026-03-01", - "last_updated": "2026-03-01", + "knowledge": "2024-12", + "release_date": "2025-01-20", + "last_updated": "2025-01-20", "modalities": { "input": ["text"], "output": ["text"] @@ -116637,104 +172302,137 @@ "open_weights": true, "limit": { "context": 128000, - "output": 16384 + "output": 32000 }, "cost": { - "input": 0.4, - "output": 1.8 + "input": 0, + "output": 0 } }, - "qwen3.5-9b": { - "id": "qwen3.5-9b", - "name": "Qwen3.5-9B", + "qwen3-vl-plus": { + "id": "qwen3-vl-plus", + "name": "Qwen3-VL-Plus", + "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks", "family": "qwen", "attachment": true, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2026-02-01", - "last_updated": "2026-02-01", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text", "image"], "output": ["text"] }, + "open_weights": false, + "limit": { + "context": 256000, + "output": 32000 + }, + "cost": { + "input": 0, + "output": 0 + } + }, + "qwen3-235b-a22b-instruct": { + "id": "qwen3-235b-a22b-instruct", + "name": "Qwen3-235B-A22B-Instruct", + "description": "Qwen instruction model for multilingual chat, reasoning, and tool use", + "family": "qwen", + "attachment": false, + "reasoning": false, + "tool_call": true, + "temperature": true, + "knowledge": "2025-04", + "release_date": "2025-07-01", + "last_updated": "2025-07-01", + "modalities": { + "input": ["text"], + "output": ["text"] + }, "open_weights": true, "limit": { - "context": 262144, - "output": 8192 + "context": 256000, + "output": 64000 }, "cost": { - "input": 0.15, - "output": 0.6 + "input": 0, + "output": 0 } }, - "gpt-oss-120b": { - "id": "gpt-oss-120b", - "name": "GPT-OSS-120B", - "family": "gpt-oss", + "kimi-k2-0905": { + "id": "kimi-k2-0905", + "name": "Kimi-K2-0905", + "description": "Kimi model for long-context chat, coding, and agentic reasoning", + "family": "kimi-k2", "attachment": false, - "reasoning": true, + "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-08-05", - "last_updated": "2025-08-05", + "knowledge": "2024-12", + "release_date": "2025-09-05", + "last_updated": "2025-09-05", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 128000, - "output": 16384 + "context": 256000, + "output": 64000 }, "cost": { - "input": 1, - "output": 4.2 + "input": 0, + "output": 0 } }, - "llama-3.1-8b-instruct": { - "id": "llama-3.1-8b-instruct", - "name": "Llama 3.1 8B Instruct", - "family": "llama", + "deepseek-v3.2": { + "id": "deepseek-v3.2", + "name": "DeepSeek-V3.2-Exp", + "description": "DeepSeek chat model for instruction following, coding, and analysis", + "family": "deepseek", "attachment": false, "reasoning": false, "tool_call": true, "temperature": true, - "release_date": "2025-04-07", - "last_updated": "2025-04-07", + "knowledge": "2024-12", + "release_date": "2025-01-01", + "last_updated": "2025-01-01", "modalities": { "input": ["text"], "output": ["text"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 120000, - "output": 120000 + "context": 128000, + "output": 64000 }, "cost": { - "input": 0.05, - "output": 0.25 + "input": 0, + "output": 0 } } } }, - "xiaomi-token-plan-ams": { - "id": "xiaomi-token-plan-ams", + "xiaomi-token-plan-sgp": { + "id": "xiaomi-token-plan-sgp", "env": ["XIAOMI_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://token-plan-ams.xiaomimimo.com/v1", - "name": "Xiaomi Token Plan (Europe)", + "api": "https://token-plan-sgp.xiaomimimo.com/v1", + "name": "Xiaomi Token Plan (Singapore)", "doc": "https://platform.xiaomimimo.com/#/docs", "models": { - "mimo-v2-tts": { - "id": "mimo-v2-tts", - "name": "MiMo-V2-TTS", + "mimo-v2.5-tts": { + "id": "mimo-v2.5-tts", + "name": "MiMo-V2.5-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "family": "mimo", "attachment": false, "reasoning": false, "tool_call": false, - "release_date": "2026-03-18", - "last_updated": "2026-03-18", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["audio"] @@ -116742,35 +172440,41 @@ "open_weights": true, "limit": { "context": 8192, - "output": 16384 + "output": 8192 }, "cost": { "input": 0, "output": 0 } }, - "mimo-v2-flash": { - "id": "mimo-v2-flash", - "name": "MiMo-V2-Flash", + "mimo-v2.5-pro": { + "id": "mimo-v2.5-pro", + "name": "MiMo-V2.5-Pro", + "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution", "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" }, "temperature": true, - "knowledge": "2024-12-01", - "release_date": "2025-12-16", - "last_updated": "2026-02-04", + "knowledge": "2024-12", + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], "output": ["text"] }, "open_weights": true, "limit": { - "context": 262144, - "output": 65536 + "context": 1048576, + "output": 131072 }, "cost": { "input": 0, @@ -116781,9 +172485,15 @@ "mimo-v2-pro": { "id": "mimo-v2-pro", "name": "MiMo-V2-Pro", + "description": "Earlier MiMo Pro model for multimodal agents, reasoning, and code tasks", "family": "mimo", "attachment": false, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -116801,76 +172511,49 @@ "context": 1048576, "output": 131072 }, + "status": "deprecated", "cost": { "input": 0, "output": 0, "cache_read": 0 } }, - "mimo-v2.5": { - "id": "mimo-v2.5", - "name": "MiMo-V2.5", - "family": "mimo", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", - "release_date": "2026-04-22", - "last_updated": "2026-04-22", - "modalities": { - "input": ["text", "image", "audio", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 1048576, - "output": 131072 - }, - "cost": { - "input": 0, - "output": 0, - "cache_read": 0 - } - }, - "mimo-v2-omni": { - "id": "mimo-v2-omni", - "name": "MiMo-V2-Omni", + "mimo-v2-tts": { + "id": "mimo-v2-tts", + "name": "MiMo-V2-TTS", + "description": "Speech generation model for controllable voice, narration, and audio delivery", "family": "mimo", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2024-12", + "attachment": false, + "reasoning": false, + "tool_call": false, "release_date": "2026-03-18", "last_updated": "2026-03-18", "modalities": { - "input": ["text", "image", "audio", "video", "pdf"], - "output": ["text"] + "input": ["text"], + "output": ["audio"] }, - "open_weights": false, + "open_weights": true, "limit": { - "context": 262144, - "output": 131072 + "context": 8192, + "output": 8192 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0 + "output": 0 } }, - "mimo-v2.5-pro": { - "id": "mimo-v2.5-pro", - "name": "MiMo-V2.5-Pro", + "mimo-v2.5": { + "id": "mimo-v2.5", + "name": "MiMo-V2.5", + "description": "Open MiMo model for multimodal coding agents and long-context automation", "family": "mimo", - "attachment": false, + "attachment": true, "reasoning": true, + "reasoning_options": [ + { + "type": "toggle" + } + ], "tool_call": true, "interleaved": { "field": "reasoning_content" @@ -116880,7 +172563,7 @@ "release_date": "2026-04-22", "last_updated": "2026-04-22", "modalities": { - "input": ["text"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": true, @@ -116893,405 +172576,115 @@ "output": 0, "cache_read": 0 } - } - } - }, - "zhipuai": { - "id": "zhipuai", - "env": ["ZHIPU_API_KEY"], - "npm": "@ai-sdk/openai-compatible", - "api": "https://open.bigmodel.cn/api/paas/v4", - "name": "Zhipu AI", - "doc": "https://docs.z.ai/guides/overview/pricing", - "models": { - "glm-5v-turbo": { - "id": "glm-5v-turbo", - "name": "GLM-5V-Turbo", - "family": "glm", - "attachment": true, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-04-01", - "last_updated": "2026-04-01", - "modalities": { - "input": ["text", "image", "video", "pdf"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 5, - "output": 22, - "cache_read": 1.2, - "cache_write": 0 - } - }, - "glm-5": { - "id": "glm-5", - "name": "GLM-5", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "release_date": "2026-02-11", - "last_updated": "2026-02-11", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 1, - "output": 3.2, - "cache_read": 0.2, - "cache_write": 0 - } - }, - "glm-5.1": { - "id": "glm-5.1", - "name": "GLM-5.1", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "structured_output": true, - "temperature": true, - "release_date": "2026-03-27", - "last_updated": "2026-03-27", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": false, - "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 6, - "output": 24, - "cache_read": 1.3, - "cache_write": 0 - } }, - "glm-4.7-flash": { - "id": "glm-4.7-flash", - "name": "GLM-4.7-Flash", - "family": "glm-flash", + "mimo-v2.5-tts-voicedesign": { + "id": "mimo-v2.5-tts-voicedesign", + "name": "MiMo-V2.5-TTS-VoiceDesign", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": true, "limit": { - "context": 200000, - "output": 131072 + "context": 8192, + "output": 8192 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 + "output": 0 } }, - "glm-4.5-flash": { - "id": "glm-4.5-flash", - "name": "GLM-4.5-Flash", - "family": "glm-flash", + "mimo-v2.5-tts-voiceclone": { + "id": "mimo-v2.5-tts-voiceclone", + "name": "MiMo-V2.5-TTS-VoiceClone", + "description": "Speech generation model for controllable voice, narration, and audio delivery", + "family": "mimo", "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", + "reasoning": false, + "tool_call": false, + "release_date": "2026-04-22", + "last_updated": "2026-04-22", "modalities": { "input": ["text"], - "output": ["text"] + "output": ["audio"] }, "open_weights": true, "limit": { - "context": 131072, - "output": 98304 + "context": 8192, + "output": 8192 }, "cost": { "input": 0, - "output": 0, - "cache_read": 0, - "cache_write": 0 - } - }, - "glm-4.6v": { - "id": "glm-4.6v", - "name": "GLM-4.6V", - "family": "glm", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-08", - "last_updated": "2025-12-08", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 128000, - "output": 32768 - }, - "cost": { - "input": 0.3, - "output": 0.9 - } - }, - "glm-4.6": { - "id": "glm-4.6", - "name": "GLM-4.6", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-09-30", - "last_updated": "2025-09-30", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 - } - }, - "glm-4.5v": { - "id": "glm-4.5v", - "name": "GLM-4.5V", - "family": "glm", - "attachment": true, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-08-11", - "last_updated": "2025-08-11", - "modalities": { - "input": ["text", "image", "video"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 64000, - "output": 16384 - }, - "cost": { - "input": 0.6, - "output": 1.8 - } - }, - "glm-4.5-air": { - "id": "glm-4.5-air", - "name": "GLM-4.5-Air", - "family": "glm-air", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 98304 - }, - "cost": { - "input": 0.2, - "output": 1.1, - "cache_read": 0.03, - "cache_write": 0 - } - }, - "glm-4.5": { - "id": "glm-4.5", - "name": "GLM-4.5", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-07-28", - "last_updated": "2025-07-28", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 131072, - "output": 98304 - }, - "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 - } - }, - "glm-4.7-flashx": { - "id": "glm-4.7-flashx", - "name": "GLM-4.7-FlashX", - "family": "glm-flash", - "attachment": false, - "reasoning": true, - "tool_call": true, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2026-01-19", - "last_updated": "2026-01-19", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 200000, - "output": 131072 - }, - "cost": { - "input": 0.07, - "output": 0.4, - "cache_read": 0.01, - "cache_write": 0 - } - }, - "glm-4.7": { - "id": "glm-4.7", - "name": "GLM-4.7", - "family": "glm", - "attachment": false, - "reasoning": true, - "tool_call": true, - "interleaved": { - "field": "reasoning_content" - }, - "temperature": true, - "knowledge": "2025-04", - "release_date": "2025-12-22", - "last_updated": "2025-12-22", - "modalities": { - "input": ["text"], - "output": ["text"] - }, - "open_weights": true, - "limit": { - "context": 204800, - "output": 131072 - }, - "cost": { - "input": 0.6, - "output": 2.2, - "cache_read": 0.11, - "cache_write": 0 + "output": 0 } } } }, - "nova": { - "id": "nova", - "env": ["NOVA_API_KEY"], + "claudinio": { + "id": "claudinio", + "env": ["CLAUDINIO_API_KEY"], "npm": "@ai-sdk/openai-compatible", - "api": "https://api.nova.amazon.com/v1", - "name": "Nova", - "doc": "https://nova.amazon.com/dev/documentation", + "api": "https://api.claudin.io/v1", + "name": "Claudinio", + "doc": "https://claudin.io", "models": { - "nova-2-lite-v1": { - "id": "nova-2-lite-v1", - "name": "Nova 2 Lite", - "family": "nova-lite", + "claudinio": { + "id": "claudinio", + "name": "Claudinio", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-12-01", - "last_updated": "2025-12-01", + "knowledge": "2026-05", + "release_date": "2026-05-12", + "last_updated": "2026-06-02", "modalities": { - "input": ["text", "image", "video", "pdf"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 256000, "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "reasoning": 0 + "input": 0.5, + "output": 2, + "cache_read": 0.15 } }, - "nova-2-pro-v1": { - "id": "nova-2-pro-v1", - "name": "Nova 2 Pro", - "family": "nova-pro", + "claudius": { + "id": "claudius", + "name": "Claudius", + "description": "Multimodal reasoning model for visual analysis, planning, and tool use", "attachment": true, "reasoning": true, + "reasoning_options": [], "tool_call": true, - "temperature": true, - "release_date": "2025-12-03", - "last_updated": "2026-01-03", + "knowledge": "2026-05", + "release_date": "2026-05-12", + "last_updated": "2026-05-12", "modalities": { - "input": ["text", "image", "video", "pdf"], + "input": ["text", "image", "audio", "video"], "output": ["text"] }, "open_weights": false, "limit": { - "context": 1000000, + "context": 256000, "output": 64000 }, "cost": { - "input": 0, - "output": 0, - "reasoning": 0 + "input": 3, + "output": 8, + "cache_read": 0.9 } } } diff --git a/packages/plugin/package.json b/packages/plugin/package.json index dd2200124198..1c23a3cebbb8 100644 --- a/packages/plugin/package.json +++ b/packages/plugin/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/plugin", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "scripts": { diff --git a/packages/sdk/js/package.json b/packages/sdk/js/package.json index 81e184968f18..555d4dd70e5d 100644 --- a/packages/sdk/js/package.json +++ b/packages/sdk/js/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/sdk", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "scripts": { diff --git a/packages/server/package.json b/packages/server/package.json index 9872b789c3ea..4111383fb47e 100644 --- a/packages/server/package.json +++ b/packages/server/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/server", - "version": "1.17.18", + "version": "1.17.19", "private": true, "type": "module", "license": "MIT", diff --git a/packages/session-ui/package.json b/packages/session-ui/package.json index cf3be61f3650..d32280dce68c 100644 --- a/packages/session-ui/package.json +++ b/packages/session-ui/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/session-ui", - "version": "1.17.18", + "version": "1.17.19", "private": true, "type": "module", "license": "MIT", diff --git a/packages/session-ui/src/components/message-file.test.ts b/packages/session-ui/src/components/message-file.test.ts index 7bdf007631c2..3882be027ef1 100644 --- a/packages/session-ui/src/components/message-file.test.ts +++ b/packages/session-ui/src/components/message-file.test.ts @@ -1,6 +1,6 @@ import { describe, expect, test } from "bun:test" import type { FilePart } from "@opencode-ai/sdk/v2" -import { attached, inline, kind } from "./message-file" +import { attached, inline, kind, typeLabel } from "./message-file" function file(part: Partial = {}): FilePart { return { @@ -52,4 +52,14 @@ describe("message-file", () => { expect(kind(file({ mime: "image/png" }))).toBe("image") expect(kind(file({ mime: "application/pdf" }))).toBe("file") }) + + test("labels attachment types from the basename extension", () => { + expect(typeLabel("list.md", "text/plain")).toBe("Markdown") + expect(typeLabel("/repo/src/main.ts", "text/plain")).toBe("TypeScript") + expect(typeLabel("/tmp/report.pdf", "application/pdf")).toBe("PDF") + expect(typeLabel("notes.xyz", "text/plain")).toBe("XYZ") + expect(typeLabel("/home/user/my.project/Makefile", "text/plain")).toBe("File") + expect(typeLabel(".gitignore", "text/plain")).toBe("File") + expect(typeLabel("/repo/.env", "text/plain")).toBe("File") + }) }) diff --git a/packages/session-ui/src/components/message-file.ts b/packages/session-ui/src/components/message-file.ts index ecc74569022e..81ce97827f97 100644 --- a/packages/session-ui/src/components/message-file.ts +++ b/packages/session-ui/src/components/message-file.ts @@ -1,3 +1,5 @@ +import { bundledLanguagesInfo } from "shiki" +import { getFilename } from "@opencode-ai/core/util/path" import type { FilePart } from "@opencode-ai/sdk/v2" export function attached(part: FilePart) { @@ -12,3 +14,22 @@ export function inline(part: FilePart) { export function kind(part: FilePart) { return part.mime.startsWith("image/") ? "image" : "file" } + +// language metadata only; grammars stay behind shiki's lazy imports +const LANGUAGE_NAMES = new Map( + bundledLanguagesInfo.flatMap((info) => + [info.id, ...(info.aliases ?? [])].map((alias) => [alias, info.name] as [string, string]), + ), +) + +// attachments carry text/plain for all text files, so the label comes from the extension; +// filename may be an absolute path, so extract the basename before looking for one +export function typeLabel(filename: string, mime: string) { + if (mime === "application/pdf") return "PDF" + const base = getFilename(filename) + // idx 0 is a dotfile like .gitignore, not an extension + const idx = base.lastIndexOf(".") + const suffix = idx <= 0 ? "" : base.slice(idx + 1).toLowerCase() + if (!suffix) return "File" + return LANGUAGE_NAMES.get(suffix) ?? suffix.toUpperCase() +} diff --git a/packages/session-ui/src/components/message-part.css b/packages/session-ui/src/components/message-part.css index 88779a3c6c82..e298ba7abc03 100644 --- a/packages/session-ui/src/components/message-part.css +++ b/packages/session-ui/src/components/message-part.css @@ -57,8 +57,20 @@ } &[data-type="image"] { - width: 48px; - height: 48px; + position: relative; + width: 58px; + height: 46px; + border: none; + + /* inset box-shadows do not paint over content, so the hairline is an overlay */ + &::after { + content: ""; + position: absolute; + inset: 0; + border-radius: inherit; + box-shadow: inset 0 0 0 0.5px var(--v2-border-border-base); + pointer-events: none; + } } &[data-type="file"] { @@ -1323,6 +1335,16 @@ body:not([data-new-layout]) { &:hover { border-color: var(--border-strong-base); } + + &[data-type="image"] { + width: 48px; + height: 48px; + border: 1px solid var(--border-weak-base); + + &::after { + content: none; + } + } } [data-slot="user-message-attachment-name"] { diff --git a/packages/session-ui/src/components/message-part.tsx b/packages/session-ui/src/components/message-part.tsx index 1ba6be82c8cb..6d1d399e6280 100644 --- a/packages/session-ui/src/components/message-part.tsx +++ b/packages/session-ui/src/components/message-part.tsx @@ -47,6 +47,8 @@ import { DiffChanges } from "@opencode-ai/ui/diff-changes" import { Markdown } from "./markdown" import { ImagePreview } from "@opencode-ai/ui/image-preview" import { getDirectory as _getDirectory, getFilename } from "@opencode-ai/core/util/path" +import { AttachmentCardV2 } from "../v2/components/attachment-card-v2" +import { CommentCardV2 } from "../v2/components/comment-card-v2" import { checksum } from "@opencode-ai/core/util/encode" import { Tooltip } from "@opencode-ai/ui/tooltip" import { IconButton } from "@opencode-ai/ui/icon-button" @@ -60,7 +62,7 @@ import { ToolStatusTitle } from "./tool-status-title" import { patchFiles } from "./apply-patch-file" import { animate } from "motion" import { useLocation } from "@solidjs/router" -import { attached, inline, kind } from "./message-file" +import { attached, inline, kind, typeLabel } from "./message-file" import { readPartText } from "./message-part-text" import { SessionProgressIndicatorV2 } from "../v2/components/session-progress-indicator-v2" @@ -167,6 +169,7 @@ export interface MessageProps { showAssistantCopyPartID?: string | null showReasoningSummaries?: boolean useV2Actions?: boolean + comments?: UserMessageComment[] } export type SessionAction = (input: { sessionID: string; messageID: string }) => Promise | void @@ -174,6 +177,16 @@ export type SessionAction = (input: { sessionID: string; messageID: string }) => export type UserActions = { fork?: SessionAction revert?: SessionAction + openAttachment?: (file: FilePart) => void +} + +export type UserMessageComment = { + path: string + comment: string + selection?: { + startLine: number + endLine: number + } } export interface MessagePartProps { @@ -946,6 +959,7 @@ export function Message(props: MessageProps) { parts={props.parts} actions={props.actions} useV2Actions={props.useV2Actions} + comments={props.comments} /> )} @@ -1163,6 +1177,7 @@ export function UserMessageDisplay(props: { parts: PartType[] actions?: UserActions useV2Actions?: boolean + comments?: UserMessageComment[] }) { const data = useData() const dialog = useDialog() @@ -1184,6 +1199,8 @@ export function UserMessageDisplay(props: { const attachments = createMemo(() => files().filter(attached)) + const messageComments = createMemo(() => (newLayout() ? (props.comments ?? []) : [])) + const inlineFiles = createMemo(() => files().filter(inline)) const agents = createMemo(() => (props.parts?.filter((p) => p.type === "agent") as AgentPart[]) ?? []) @@ -1240,35 +1257,59 @@ export function UserMessageDisplay(props: { return (
- 0}> + 0 || messageComments().length > 0}>
+ + {(comment) => ( + + )} + {(file) => { const type = kind(file) const name = file.filename ?? i18n.t("ui.message.attachment.alt") return ( -
{ - if (type === "image") openImagePreview(file.url, name) - }} + { + if (type === "image") openImagePreview(file.url, name) + }} + > + + + {name} +
+ } + > + {name} + +
+ } > - - - {name} -
- } + props.actions?.openAttachment?.(file)} > - {name} - -
+ {typeLabel(name, file.mime)} + + ) }} diff --git a/packages/session-ui/src/v2/components/attachment-card-v2.css b/packages/session-ui/src/v2/components/attachment-card-v2.css new file mode 100644 index 000000000000..1a19dc7b5edf --- /dev/null +++ b/packages/session-ui/src/v2/components/attachment-card-v2.css @@ -0,0 +1,58 @@ +[data-component="attachment-card-v2"] { + display: flex; + flex-direction: column; + gap: 6px; + box-sizing: border-box; + width: 160px; + min-width: 160px; + max-width: 160px; + padding: 8px; + border-radius: 6px; + background: var(--v2-overlay-simple-overlay-hover); + box-shadow: inset 0 0 0 0.5px var(--v2-border-border-base); + cursor: default; + + &[data-active] { + box-shadow: inset 0 0 0 0.5px var(--v2-border-border-strong); + } + + &[data-clickable] { + cursor: pointer; + } + + [data-slot="attachment-card-v2-title"], + [data-slot="attachment-card-v2-subtitle"] { + max-width: 100%; + font-family: var(--v2-font-family-sans, "Inter", sans-serif); + font-style: normal; + font-size: 11px; + line-height: 12px; + letter-spacing: 0.05px; + font-variation-settings: "slnt" 0; + } + + [data-slot="attachment-card-v2-title"] { + overflow: hidden; + text-overflow: ellipsis; + white-space: nowrap; + font-weight: 530; + color: var(--v2-text-text-base); + } + + [data-slot="attachment-card-v2-subtitle"] { + display: flex; + align-items: center; + gap: 4px; + min-width: 0; + overflow: hidden; + white-space: nowrap; + font-weight: 440; + color: var(--v2-text-text-muted); + + [data-component="file-icon"] { + width: 12px; + height: 12px; + flex: none; + } + } +} diff --git a/packages/session-ui/src/v2/components/attachment-card-v2.tsx b/packages/session-ui/src/v2/components/attachment-card-v2.tsx new file mode 100644 index 000000000000..2e896383a52b --- /dev/null +++ b/packages/session-ui/src/v2/components/attachment-card-v2.tsx @@ -0,0 +1,26 @@ +import type { JSX } from "solid-js" +import "./attachment-card-v2.css" + +/** Shared 160px two-line card used by v2 file and comment attachments in the composer and timeline. */ +export function AttachmentCardV2(props: { + title: string + active?: boolean + clickable?: boolean + /** native title attribute */ + hover?: string + onClick?: () => void + children: JSX.Element +}) { + return ( +
props.onClick?.()} + > + {props.title} + {props.children} +
+ ) +} diff --git a/packages/session-ui/src/v2/components/comment-card-v2.tsx b/packages/session-ui/src/v2/components/comment-card-v2.tsx new file mode 100644 index 000000000000..fa2fe08c329d --- /dev/null +++ b/packages/session-ui/src/v2/components/comment-card-v2.tsx @@ -0,0 +1,27 @@ +import { Show } from "solid-js" +import { FileIcon } from "@opencode-ai/ui/file-icon" +import { getFilenameTruncated } from "@opencode-ai/core/util/path" +import { AttachmentCardV2 } from "./attachment-card-v2" + +export function CommentCardV2(props: { + comment: string + path: string + selection?: { startLine: number; endLine: number } + active?: boolean + title?: string + onClick?: () => void +}) { + return ( + + + + {getFilenameTruncated(props.path, 14)} + + {(sel) => + sel().startLine === sel().endLine ? `:${sel().startLine}` : `:${sel().startLine}-${sel().endLine}` + } + + + + ) +} diff --git a/packages/session-ui/src/v2/components/line-comment-annotations-v2.tsx b/packages/session-ui/src/v2/components/line-comment-annotations-v2.tsx index 4e8ed8673220..cd27a1ecc7b5 100644 --- a/packages/session-ui/src/v2/components/line-comment-annotations-v2.tsx +++ b/packages/session-ui/src/v2/components/line-comment-annotations-v2.tsx @@ -11,6 +11,7 @@ import { import { useI18n } from "@opencode-ai/ui/context/i18n" import { cloneSelectedLineRange, formatSelectedLineLabel } from "../../pierre/selection-bridge" import { LineCommentEditorV2, LineCommentV2 } from "@opencode-ai/ui/v2/line-comment-v2" +import type { LineCommentEditorV2Mention } from "@opencode-ai/ui/v2/line-comment-v2" type LineCommentControllerV2Props = { comments: Accessor @@ -23,6 +24,7 @@ type LineCommentControllerV2Props = { onDelete?: (comment: T) => void renderCommentActions?: (comment: T, controls: { edit: VoidFunction; remove: VoidFunction }) => JSX.Element editSubmitLabel?: string + mention?: LineCommentEditorV2Mention } type CommentProps = { @@ -43,6 +45,7 @@ type DraftProps = { onSubmit: (value: string) => void cancelLabel?: string submitLabel?: string + mention?: LineCommentEditorV2Mention } function lineCommentElementV2(view: Accessor) { @@ -70,6 +73,7 @@ function lineCommentElementV2(view: Accessor) { onSubmit={view().editor!.onSubmit} cancelLabel={view().editor!.cancelLabel} submitLabel={view().editor!.submitLabel} + mention={view().editor!.mention} />
@@ -87,6 +91,7 @@ function lineCommentDraftElementV2(view: Accessor) { onSubmit={view().onSubmit} cancelLabel={view().cancelLabel} submitLabel={view().submitLabel} + mention={view().mention} />
) @@ -142,6 +147,7 @@ export function createLineCommentControllerV2(props: }, cancelLabel: i18n.t("ui.lineComment.cancel"), submitLabel: props.editSubmitLabel, + mention: props.mention, } : undefined }, @@ -165,6 +171,7 @@ export function createLineCommentControllerV2(props: }, cancelLabel: i18n.t("ui.lineComment.cancel"), submitLabel: i18n.t("ui.lineComment.submit"), + mention: props.mention, }), }) @@ -202,6 +209,7 @@ export function createLineCommentControllerV2(props: } return { + note, annotations, renderAnnotation, renderGutterUtility, diff --git a/packages/session-ui/src/v2/components/session-review-v2.css b/packages/session-ui/src/v2/components/session-review-v2.css index c42fd20d3b2a..9db2ba580c33 100644 --- a/packages/session-ui/src/v2/components/session-review-v2.css +++ b/packages/session-ui/src/v2/components/session-review-v2.css @@ -4,7 +4,7 @@ height: 100%; min-height: 0; overflow: hidden; - background: var(--background-stronger, var(--v2-background-bg-base)); + background: var(--v2-background-bg-base); } [data-component="session-review-v2"] [data-slot="session-review-v2-body"] { @@ -31,7 +31,7 @@ min-height: 0; overflow: hidden; border-right: 1px solid var(--border-weaker-base, var(--v2-border-border-weak)); - background: var(--background-stronger, var(--v2-background-bg-base)); + background: var(--v2-background-bg-base); } [data-component="session-review-v2-sidebar-root"] [data-slot="session-review-v2-sidebar"][aria-hidden="true"] { @@ -76,6 +76,11 @@ color: var(--text-strong, var(--v2-text-text-base)); } +[data-component="session-review-v2-sidebar-root"] + [data-slot="session-review-v2-sidebar-title"]:not(:has([data-component="select-v2-root"])) { + margin-left: 8px; +} + [data-component="session-review-v2-sidebar-root"] [data-slot="session-review-v2-sidebar-title"] [data-component="select-v2-root"] { @@ -97,7 +102,7 @@ width: 100%; max-width: 100%; min-width: 0; - margin-top: 2px; + margin-top: 3px; } [data-component="session-review-v2-sidebar-root"] [data-slot="session-review-v2-sidebar-tree"] { diff --git a/packages/session-ui/src/v2/components/session-review-v2.tsx b/packages/session-ui/src/v2/components/session-review-v2.tsx index 3c13c75d7bb1..6efb2b593ab5 100644 --- a/packages/session-ui/src/v2/components/session-review-v2.tsx +++ b/packages/session-ui/src/v2/components/session-review-v2.tsx @@ -26,7 +26,6 @@ export type SessionReviewV2Props = { empty?: JSX.Element sidebarOpen?: boolean sidebar?: JSX.Element - sidebarToggle?: JSX.Element activeFile?: string files: string[] onSelectFile: (file: string) => void @@ -199,7 +198,6 @@ export function SessionReviewV2(props: SessionReviewV2Props) { const toolbarStart = () => ( <> - {props.sidebarToggle}
@@ -319,7 +317,7 @@ export function SessionReviewV2(props: SessionReviewV2Props) { ) } -export function SessionReviewV2SidebarToggle(props: { opened: boolean; onToggle: () => void }) { +export function SessionReviewV2SidebarToggle(props: { opened: boolean; disabled?: boolean; onToggle: () => void }) { const i18n = useI18n() return ( @@ -330,6 +328,7 @@ export function SessionReviewV2SidebarToggle(props: { opened: boolean; onToggle: class="session-review-v2-sidebar-toggle" aria-label={i18n.t("ui.sessionReviewV2.toggleSidebar")} aria-expanded={props.opened} + disabled={props.disabled} onClick={props.onToggle} icon={} /> diff --git a/packages/slack/package.json b/packages/slack/package.json index 11bb354f0d34..7f8bcedac11d 100644 --- a/packages/slack/package.json +++ b/packages/slack/package.json @@ -1,6 +1,6 @@ { "name": "@opencode-ai/slack", - "version": "1.17.18", + "version": "1.17.19", "type": "module", "license": "MIT", "scripts": { diff --git a/packages/stats/app/package.json b/packages/stats/app/package.json index f1def44cecbb..67abf1afd942 100644 --- a/packages/stats/app/package.json +++ b/packages/stats/app/package.json @@ -1,7 +1,7 @@ { "$schema": "https://json.schemastore.org/package.json", "name": "@opencode-ai/stats-app", - "version": "1.17.18", + "version": "1.17.19", "private": true, "type": "module", "license": "MIT", diff --git a/packages/stats/app/src/component/model-compare-detail.tsx b/packages/stats/app/src/component/model-compare-detail.tsx new file mode 100644 index 000000000000..68cfcfee6882 --- /dev/null +++ b/packages/stats/app/src/component/model-compare-detail.tsx @@ -0,0 +1,1150 @@ +import "../routes/index.css" +import { Link, Meta, Title } from "@solidjs/meta" +import { ProviderIcon } from "@opencode-ai/ui/provider-icon" +import { + getStatsModelsComparisonData, + type ModelUsagePoint, + type StatsModelComparisonInput, + type StatsModelComparisonEntry, +} from "@opencode-ai/stats-core/domain/home" +import { runtime } from "@opencode-ai/stats-core/runtime" +import { createAsync, query, useParams, useSearchParams } from "@solidjs/router" +import { createEffect, createMemo, createSignal, For, onMount, Show } from "solid-js" +import { getRequestEvent } from "solid-js/web" +import { + ComparisonCardsSection, + comparisonHref, + modelRefFromCatalog, + uniqueComparisonPairs, + type ComparisonModelRef, + type ComparisonPair, +} from "../routes/compare-cards" +import { ComparisonRadar } from "../routes/compare-radar" +import { + catalogSlug, + findModelCatalogEntry, + formatCatalogLabName, + getModelCatalog, + type ModelCatalog, + type ModelCatalogEntry, +} from "../routes/model-catalog" +import { + applyThemePreference, + Footer, + getGitHubStars, + Header, + isThemePreference, + themeStorageKey, + type HeaderLink, + type ThemePreference, +} from "../routes/stats-shell" +import { + canonicalFamilyComparisonPath, + canonicalModelComparisonPath, + latestFamilyComparisonPath, + type ResolvedComparisonFamily, +} from "../lib/comparison-pages" +import { baseUrl } from "../lib/language" + +const compareHeaderLinks: readonly HeaderLink[] = [ + { href: `${import.meta.env.BASE_URL}#top-models`, label: "Top Models" }, + { href: `${import.meta.env.BASE_URL}#leaderboard`, label: "Leaderboard" }, + { href: `${import.meta.env.BASE_URL}#market-share`, label: "Market Share" }, + { href: `${import.meta.env.BASE_URL}#token-cost`, label: "Token Cost" }, + { href: `${import.meta.env.BASE_URL}#session-cost`, label: "Session Cost" }, +] +const compareFooterLinks: readonly HeaderLink[] = [ + { href: import.meta.env.BASE_URL, label: "Data Home" }, + { href: `${import.meta.env.BASE_URL}compare`, label: "Model Compare" }, + { href: `${import.meta.env.BASE_URL}#top-models`, label: "Top Models" }, + { href: `${import.meta.env.BASE_URL}#token-cost`, label: "Token Cost" }, +] +const heroLabs = [ + { lab: "deepseek", label: "DeepSeek" }, + { lab: "openai", label: "OpenAI" }, + { lab: "anthropic", label: "Anthropic" }, +] as const +const usageBarLimit = 60 +const comparisonModelLimit = 6 + +type ComparisonModel = { + name: string + lab: string + labName: string + slug: string + catalog: ModelCatalogEntry | null + stats: StatsModelComparisonEntry | null +} +type ComparisonDirection = "higher" | "lower" +type ComparisonDetailCell = { + value: string + unit?: string + href?: string + kind?: "boolean" + score?: number + trend?: number +} +type ComparisonDetailRow = { + label: string + direction?: ComparisonDirection + cells: ComparisonDetailCell[] +} +type ComparisonDetailSection = { + title: string + badge?: string + rows: ComparisonDetailRow[] + usage?: ModelUsagePoint[][] +} +type ComparisonModelRequest = { + lab: string + slug: string +} + +export type ModelCompareDetailPageProps = { + first?: { lab: string; slug: string } + second?: { lab: string; slug: string } + family?: { first: ResolvedComparisonFamily; second: ResolvedComparisonFamily } + catalog?: ModelCatalog +} + +const getComparisonData = query(async (models: StatsModelComparisonInput[]) => { + "use server" + return runtime.runPromise(getStatsModelsComparisonData(models)) +}, "getStatsModelComparisonDetailData") + +export default function ModelCompareDetailPage(props: ModelCompareDetailPageProps = {}) { + const event = getRequestEvent() + event?.response.headers.set("Cache-Control", "public, max-age=60, s-maxage=300, stale-while-revalidate=86400") + const params = useParams() + const [searchParams] = useSearchParams<{ add?: string }>() + const firstLabParam = createMemo(() => props.first?.lab ?? params.firstLab ?? "") + const firstModelParam = createMemo(() => props.first?.slug ?? params.firstModel ?? "") + const secondLabParam = createMemo(() => props.second?.lab ?? params.secondLab ?? "") + const secondModelParam = createMemo(() => props.second?.slug ?? params.secondModel ?? "") + const catalogData = createAsync(() => getModelCatalog()) + const catalog = createMemo(() => props.catalog ?? catalogData()) + const firstCatalog = createMemo(() => resolvedCatalogEntry(catalog(), firstLabParam(), firstModelParam())) + const secondCatalog = createMemo(() => resolvedCatalogEntry(catalog(), secondLabParam(), secondModelParam())) + const modelRequests = createMemo(() => { + const selected: ComparisonModelRequest[] = [ + { lab: firstLabParam(), slug: firstModelParam() }, + { lab: secondLabParam(), slug: secondModelParam() }, + ...parseAdditionalModels(searchParams.add), + ] + return selected + .filter( + (model, index) => + index < 2 || + selected.findIndex( + (candidate) => comparisonModelRequestKey(candidate) === comparisonModelRequestKey(model), + ) === index, + ) + .slice(0, comparisonModelLimit) + }) + const modelSelections = createMemo(() => + modelRequests().map((model) => ({ + ...model, + catalog: resolvedCatalogEntry(catalog(), model.lab, model.slug), + })), + ) + const stats = createAsync(() => + getComparisonData(modelRequests().map((model) => ({ provider: model.lab, model: model.slug }))), + ) + const githubStars = createAsync(() => getGitHubStars()) + const [themePreference, setThemePreference] = createSignal("system") + const [highlightBest, setHighlightBest] = createSignal(true) + const [addingModel, setAddingModel] = createSignal(false) + let comparisonHeadingScroll: HTMLDivElement | undefined + let comparisonBodyScroll: HTMLDivElement | undefined + const models = createMemo(() => + modelSelections().map((model, index) => + buildComparisonModel(model.lab, model.slug, model.catalog ?? null, stats()?.models[index] ?? null), + ), + ) + const title = createMemo(() => { + if (props.family) + return `${props.family.first.name} vs ${props.family.second.name}: ${models()[0].name} vs ${models()[1].name}` + return `${models()[0].name} vs ${models()[1].name} - Model Comparison` + }) + const description = createMemo(() => { + if (props.family) + return `Compare the latest ${props.family.first.name} and ${props.family.second.name} models: ${models()[0].name} vs ${models()[1].name}. See benchmarks, usage, price, context length, and features.` + return `Compare ${models()[0].name} and ${models()[1].name} by usage, rank, context window, output limit, cache ratio, and cost across OpenCode data.` + }) + const canonicalPath = createMemo(() => { + if (props.family) return canonicalFamilyComparisonPath(props.family.first, props.family.second) + const first = firstCatalog() + const second = secondCatalog() + const source = catalog() + if (source && first && second) + return latestFamilyComparisonPath(source, first, second) ?? canonicalModelComparisonPath(first, second) + return canonicalModelComparisonPath(comparisonCatalogEntry(models()[0]), comparisonCatalogEntry(models()[1])) + }) + const canonicalUrl = createMemo(() => new URL(canonicalPath(), baseUrl).toString()) + const detailSections = createMemo(() => buildComparisonDetailSections(models())) + const relatedPairs = createMemo(() => buildRelatedPairs(catalog(), models()[0], models()[1])) + const selectorModels = createMemo(() => + uniqueCatalogModels([...models().map(comparisonCatalogEntry), ...(catalog()?.models ?? [])]), + ) + const selectedCatalogModels = createMemo(() => models().map(comparisonCatalogEntry)) + const canAddModel = createMemo( + () => + models().length < comparisonModelLimit && + selectorModels().some((model) => !selectedCatalogModels().some((selected) => selected.id === model.id)), + ) + const navigateToModels = (next: ModelCatalogEntry[]) => { + if (typeof window === "undefined" || next.length < 2) return + window.location.href = comparisonModelsHref(next) + } + const syncComparisonScroll = (source: HTMLDivElement, target: HTMLDivElement | undefined) => { + if (target && target.scrollLeft !== source.scrollLeft) target.scrollLeft = source.scrollLeft + } + const structuredData = createMemo(() => + JSON.stringify({ + "@context": "https://schema.org", + "@type": "WebPage", + name: title(), + description: description(), + url: canonicalUrl(), + about: models().map((model) => ({ + "@type": "SoftwareApplication", + name: model.name, + applicationCategory: "AI model", + provider: model.labName, + })), + }), + ) + const updateThemePreference = (preference: ThemePreference) => { + applyThemePreference(preference) + setThemePreference(preference) + if (typeof window === "undefined") return + window.localStorage.setItem(themeStorageKey, preference) + } + + onMount(() => { + if (typeof window === "undefined") return + const preference = window.localStorage.getItem(themeStorageKey) + const nextPreference = isThemePreference(preference) ? preference : "system" + applyThemePreference(nextPreference) + setThemePreference(nextPreference) + }) + + return ( +
+ + {title()} + + 2 ? "noindex,follow" : "index,follow"} /> + + + + + + + + + + + +
+
+
+ setAddingModel(true)} + onHighlightBestChange={() => setHighlightBest(!highlightBest())} + /> + + model.id)} + label="Add model" + models={selectorModels()} + onClose={() => setAddingModel(false)} + onSelect={(model) => { + setAddingModel(false) + navigateToModels([...selectedCatalogModels(), model]) + }} + /> + + +
+
(comparisonHeadingScroll = element)} + onScroll={(event) => syncComparisonScroll(event.currentTarget, comparisonBodyScroll)} + > + +
+
(comparisonBodyScroll = element)} + onScroll={(event) => syncComparisonScroll(event.currentTarget, comparisonHeadingScroll)} + > + +
+
+ +
+
+
+
+ ) +} + +function ComparisonHero(props: { + models: readonly ComparisonModel[] + canAddModel: boolean + highlightBest: boolean + onAddModel: () => void + onHighlightBestChange: () => void +}) { + return ( +
+ +
+

model.name).join(", ")}`}> + Compare + + AI models +

+
+ + +
+
+
+ ) +} + +function HeroModelStack() { + return ( + + ) +} + +function ComparisonPairSelector(props: { catalogModels: ModelCatalogEntry[]; models: readonly ComparisonModel[] }) { + const [activeIndex, setActiveIndex] = createSignal() + const selectedModels = createMemo(() => props.models.map(comparisonCatalogEntry)) + const activeSelected = createMemo(() => { + const index = activeIndex() + if (index === undefined) return undefined + return selectedModels()[index] + }) + const blockedIds = createMemo(() => + selectedModels() + .filter((_, index) => index !== activeIndex()) + .map((model) => model.id), + ) + const navigateToSelection = (index: number, model: ModelCatalogEntry) => { + if (typeof window === "undefined") return + window.location.href = comparisonModelsHref( + selectedModels().map((selected, selectedIndex) => (selectedIndex === index ? model : selected)), + ) + } + + return ( +
+
+ + + {(selected) => ( + setActiveIndex(undefined)} + onSelect={(model) => { + const index = activeIndex() + if (index === undefined) return + setActiveIndex(undefined) + navigateToSelection(index, model) + }} + /> + )} + +
+ ) +} + +function CompareDetailSelectButton(props: { + model: ComparisonModel + label: string + column: number + last: boolean + expanded: boolean + onOpen: () => void +}) { + return ( + + ) +} + +function CompareModelSelectModal(props: { + models: ModelCatalogEntry[] + selected?: ModelCatalogEntry + blockedIds: string[] + label: string + onClose: () => void + onSelect: (model: ModelCatalogEntry) => void +}) { + let searchInput: HTMLInputElement | undefined + const [search, setSearch] = createSignal("") + const [previewId, setPreviewId] = createSignal(props.selected?.id ?? "") + const availableModels = createMemo(() => + uniqueCatalogModels([...(props.selected ? [props.selected] : []), ...props.models]).filter( + (model) => !props.blockedIds.includes(model.id), + ), + ) + const filteredModels = createMemo(() => { + const terms = search().trim().toLowerCase().split(/\s+/).filter(Boolean) + if (terms.length === 0) return availableModels() + return availableModels().filter((model) => terms.every((term) => modelSearchText(model).includes(term))) + }) + const preview = createMemo( + () => filteredModels().find((model) => model.id === previewId()) ?? filteredModels()[0] ?? availableModels()[0], + ) + + createEffect(() => { + const models = filteredModels() + if (models.some((model) => model.id === previewId())) return + const selected = + props.selected && models.some((model) => model.id === props.selected?.id) ? props.selected.id : undefined + setPreviewId(selected ?? models[0]?.id ?? "") + }) + + createEffect(() => { + if (typeof window === "undefined") return + window.requestAnimationFrame(() => searchInput?.focus()) + }) + + return ( +
{ + if (event.key !== "Escape") return + props.onClose() + }} + > +
event.stopPropagation()} + > +
+ +
+ 0} + fallback={ +
+ No models found + Try another search. +
+ } + > + + {(model) => ( + + )} + +
+
+
+ +
+ ) +} + +function CompareModelDetail(props: { model: ModelCatalogEntry }) { + return ( + + ) +} + +function CompareModelFact(props: { label: string; value: string; href?: string }) { + return ( +
+
{props.label}
+
+ + {(href) => {props.value}} + +
+
+ ) +} + +function ComparisonDetailMatrix(props: { + sections: ComparisonDetailSection[] + highlightBest: boolean + modelCount: number +}) { + return ( +
+
+ + {(section) => ( +
+ +
+ {section.title} + {(badge) => {badge()}} +
+ + {(_, index) => ( +
+ )} +
+
+
+ ) +} + +function ComparisonDetailSpacer(props: { modelCount: number }) { + return ( + <> +