Repository navigation
feat(providers): add GPT-6 Sol and Luna ahead of every GPT-5.6 Sol provider except Cursor - #5584
Conversation
…ovider except Cursor OpenAI released GPT-6 Sol and Luna on 2026-09-22 (API list price: Sol $2 / $0.20 cached / $10, Luna $0.10 / $0.01 cached / $0.50). Most providers had not published them on 2026-09-23; the rows are added preemptively, mirroring each provider's GPT-5.6 Sol/Luna entry, the same way Opus 5.5 was added in #5581. Cursor is deliberately untouched. - Metadata snapshot + regenerated table: gpt-6-sol / gpt-6-luna for amazon-bedrock, cloudflare-ai-gateway, github-copilot, kilo, openai, openai-codex, opencode-zen, openrouter, vercel-ai-gateway and zenmux, at list price (fields a provider bills at 0, such as Kilo's subscription rows, stay 0). opencode-go has no GPT-5.6 Sol and is skipped. - Registry: OpenAI API key (1,050,000 / 922,000 / 128K and low..max, mirroring gpt-6-astra's API seed), OpenRouter models/windows/service tier, GitHub Copilot models and Responses wire, BizRouter, DigitalOcean, Devin/Devin CLI seeds. - Kiro, CodeBuddy and Devin static rosters, with Kiro's native reasoning field. - Price overlays: verified OpenAI API rows, derived rows for ChatGPT login and Devin; long-context tier rows for both.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Navigate logical layers of code changes, visualize relationships, and explore their blast radius. Note Currently processing new changes in this PR. This may take a few minutes, please wait... ⚙️ Run configurationConfiguration used: Repository: lidge-jun/opencodex/.coderabbit.yaml Review profile: ASSERTIVE Plan: Advanced Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (13)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
✅ Deterministic PR hygiene checks passed. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: bfb8fede7b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| "gpt-6-sol", | ||
| "gpt-6-luna", |
There was a problem hiding this comment.
For any GPT-6 Sol or Luna request exceeding 272K input tokens, resolveContextTier now doubles input/cache costs and multiplies output costs by 1.5. However, the added pricing source documents only the base rates, the registry comment says the API window is not yet published, and the inherited verifiedAt: "2026-09-05" predates the models' stated September 22 release. This will overstate usage costs unless these models actually use the GPT-5.6 long-context schedule; omit them until that schedule is published or give them independently sourced tier rows.
Useful? React with 👍 / 👎.
| "cacheRead": 0.01, | ||
| "cacheWrite": 0.125 | ||
| }, | ||
| "contextWindow": 373000, |
There was a problem hiding this comment.
This bundled OpenRouter row declares a 373,000-token window, while the same change declares 1,050,000 for this exact model in OPENROUTER_GPT56_CONTEXT_WINDOWS. Because model-hints.ts passes the bundled value through clampObservedModelLimits, the lower 373K value wins and auto-compaction begins around 336K instead of using the registry's intended 1.05M window; Luna has the same conflict. Keep this provider fact in one canonical source or make the two values agree.
AGENTS.md reference: src/AGENTS.md:L18-L18
Useful? React with 👍 / 👎.
…ovider except Cursor (lidge-jun#5584) OpenAI released GPT-6 Sol and Luna on 2026-09-22 (API list price: Sol $2 / $0.20 cached / $10, Luna $0.10 / $0.01 cached / $0.50). Most providers had not published them on 2026-09-23; the rows are added preemptively, mirroring each provider's GPT-5.6 Sol/Luna entry, the same way Opus 5.5 was added in lidge-jun#5581. Cursor is deliberately untouched. - Metadata snapshot + regenerated table: gpt-6-sol / gpt-6-luna for amazon-bedrock, cloudflare-ai-gateway, github-copilot, kilo, openai, openai-codex, opencode-zen, openrouter, vercel-ai-gateway and zenmux, at list price (fields a provider bills at 0, such as Kilo's subscription rows, stay 0). opencode-go has no GPT-5.6 Sol and is skipped. - Registry: OpenAI API key (1,050,000 / 922,000 / 128K and low..max, mirroring gpt-6-astra's API seed), OpenRouter models/windows/service tier, GitHub Copilot models and Responses wire, BizRouter, DigitalOcean, Devin/Devin CLI seeds. - Kiro, CodeBuddy and Devin static rosters, with Kiro's native reasoning field. - Price overlays: verified OpenAI API rows, derived rows for ChatGPT login and Devin; long-context tier rows for both.
1 participant
Summary
OpenAI released GPT-6 Sol and GPT-6 Luna on 2026-09-22. API list prices per 1M tokens:
Most providers that carry GPT-5.6 Sol had not published the new models on 2026-09-23. This PR adds them preemptively, mirroring each provider's GPT-5.6 Sol/Luna entry, the same way Opus 5.5 was added ahead of providers in #5581. Cursor is deliberately untouched. The native ChatGPT/Codex rows already landed in #5580.
Metadata snapshot and regenerated table.
gpt-6-solandgpt-6-lunaare added for these providers, cloned from their GPT-5.6 rows:openai.gpt-6-*)openai/gpt-6-*)Priced fields get the list price, and fields a provider bills at 0 stay 0 (Kilo's subscription rows, Cloudflare's cache write). Several reseller GPT-5.6 Luna rows still carry pre-cut prices, so scaling from them would have produced $0.5 / $2.5. opencode-go lists no GPT-5.6 Sol and is skipped.
Registry.
gpt-6-astra's API seed until OpenAI publishes the API window.Static rosters. Kiro (272K, native
reasoningeffort field), CodeBuddy (mirrors GPT-5.6 Sol/Luna) and Devin (1M).Price overlays.
Kiro, CodeBuddy and Devin rosters are static, so these models now appear there, and calls fail upstream until each provider ships them. Snapshot-only providers are unaffected until the model shows up in their live list.
Verification
bun run typecheck: exit 0 on this head.tests/providers/kiro/kiro-adapter.test.tsstays at 2050 lines.Checklist