[litellm-agent] Staging → litellm_internal_staging (5/19/2026) by oss-pr-review-agent-shin[bot] · Pull Request #28207 · BerriAI/litellm

oss-pr-review-agent-shin · 2026-05-19T02:38:19Z

Merged PRs (1)

#	Title
#27700	feat: add Xiaomi MiMo-V2.5-Pro and MiMo-V2.5 OpenRouter model entries

Auto-updated by litellm-agent on each merge.

…#27700) Squash-merged by litellm-agent from TorvaldUtne's PR.

oss-pr-review-agent-shin · 2026-05-19T02:38:20Z

@greptile please review

greptile-apps · 2026-05-19T02:40:15Z

Greptile Summary

This PR merges one feature PR (#27700) that adds Xiaomi MiMo-V2.5 and MiMo-V2.5-Pro OpenRouter entries alongside a bug fix that makes the Anthropic transformation accept dict-shaped reasoning_effort values produced by the OpenAI Responses→Chat bridge.

Anthropic transformation fix: Removes the hard isinstance(value, str) guard on reasoning_effort and coerces {"effort": "low", "summary": "concise"} dicts to a plain string before mapping, preventing silent disabling of extended thinking when the Responses bridge is in use. Three parametrized test suites cover string, dict-with-effort, dict-with-effort-and-summary, and malformed-dict shapes across adaptive (4.6+) and non-adaptive models.
Model pricing entries: Adds gemini-3.1-flash-lite for four prefixes (bare, gemini/, openrouter/, vertex_ai/) and the two new Xiaomi MiMo-V2.5 variants; MiMo-V2.5 multimodal flags (vision, audio, video) match OpenRouter's published description of it as a "native omnimodal model".
UI tweak: ToolTestPanel.tsx trims leading/trailing whitespace from string inputs before type conversion.

Confidence Score: 5/5

The changes are safe to merge: the Anthropic transformation fix is narrowly scoped and well-covered by new unit tests, and the pricing additions are consistent with published OpenRouter data.

The transformation change is a focused coercion fix with thorough parametrized tests for every valid and invalid shape. Pricing entries for the new Gemini and MiMo models are confirmed against OpenRouter's published pages. No logic paths are bypassed, no auth or critical-path code is touched.

No files require special attention.

Important Files Changed

Filename	Overview
litellm/llms/anthropic/chat/transformation.py	Removes `isinstance(value, str)` guard on `reasoning_effort` and adds dict-shape normalization so the OpenAI Responses→Chat bridge no longer silently drops extended thinking
tests/test_litellm/llms/anthropic/chat/test_anthropic_chat_transformation.py	Adds three parametrized test functions covering string-shape, dict-with-effort, dict-with-effort-and-summary, and malformed-dict cases for both adaptive (4.6+) and non-adaptive models — all properly mocked, no network calls
tests/test_litellm/test_cost_calculator.py	Adds a regression test for the new `openrouter/google/gemini-3.1-flash-lite` pricing entry; uses `LITELLM_LOCAL_MODEL_COST_MAP=True` so no network calls are made
model_prices_and_context_window.json	Adds gemini-3.1-flash-lite entries for four prefixes (bare, gemini/, openrouter/, vertex_ai/), adds openrouter/xiaomi/mimo-v2.5-pro and mimo-v2.5 entries (pricing confirmed via OpenRouter), and updates mimo-v2-flash pricing/caching fields
litellm/model_prices_and_context_window_backup.json	Mirror of the main pricing file kept in sync; same additions as model_prices_and_context_window.json
litellm/proxy/_lazy_openapi_snapshot.json	Updates the POST /agents example in the OpenAPI snapshot: corrects the URL to `/v1/agents` and removes the unnecessary wrapping `agent` object from the request body
ui/litellm-dashboard/src/components/mcp_tools/ToolTestPanel.tsx	Adds `.trim()` to string inputs before type conversion so leading/trailing whitespace is stripped before submission; correctly handles non-string types by conditionally applying trim

_{Reviews (2): Last reviewed commit: "feat: add pricing entry for openrouter/g..." | Re-trigger Greptile}

greptile-apps · 2026-05-19T02:40:20Z

    "openrouter/xiaomi/mimo-v2-flash": {
-        "input_cost_per_token": 9e-08,
-        "output_cost_per_token": 2.9e-07,
+        "input_cost_per_token": 1e-07,
+        "output_cost_per_token": 3e-07,
        "cache_creation_input_token_cost": 0.0,
-        "cache_read_input_token_cost": 0.0,
+        "cache_read_input_token_cost": 1e-08,


The updated input_cost_per_token and output_cost_per_token for mimo-v2-flash don't match the prices listed on OpenRouter's official model page, which still shows $0.09/M input and $0.29/M output — the pre-existing values. Bumping them to $0.10/M and $0.30/M will cause litellm to overestimate costs for every request to this model.

Suggested change

"openrouter/xiaomi/mimo-v2-flash": {

"input_cost_per_token": 9e-08,

"output_cost_per_token": 2.9e-07,

"input_cost_per_token": 1e-07,

"output_cost_per_token": 3e-07,

"cache_creation_input_token_cost": 0.0,

"cache_read_input_token_cost": 0.0,

"cache_read_input_token_cost": 1e-08,

"openrouter/xiaomi/mimo-v2-flash": {

"input_cost_per_token": 9e-08,

"output_cost_per_token": 2.9e-07,

"cache_creation_input_token_cost": 0.0,

"cache_read_input_token_cost": 1e-08,

codecov · 2026-05-19T02:41:23Z

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai>

CLAassistant · 2026-05-19T03:07:53Z

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
4 out of 6 committers have signed the CLA.

✅ TorvaldUtne
✅ mubashir1osmani
✅ ro31337
✅ cwang-otto
❌ oss-agent-shin
❌ IshaMeera
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

* feat(model_prices): add gemini-3.1-flash-lite pricing with standard/batch/flex/priority tiers * fix pricing * add service tier --------- Co-authored-by: shin-berri <shin-laptop@berri.ai>

@greptile-apps

…dge (#28201) * fix(anthropic): accept dict-shape reasoning_effort from Responses bridge Issue #28196 — the Responses->Chat parser (transformation.py:184-200) keeps the full dict as reasoning_effort when summary is set; that branch was added in #25359. But the Anthropic transformation here still guarded on isinstance(value, str), silently dropping the param. Result: callers using the standard Reasoning(effort, summary) OpenAI-shaped object on Anthropic lose thinking entirely (0 reasoning_tokens, no thinking_blocks). Coerce dict -> string before mapping. Same shape tolerance that gpt_5_transformation._normalize_reasoning_effort_for_chat_completion already implements. summary is irrelevant for Anthropic's thinking_blocks. Adds two regression tests: one parametrized over string + dict shapes (with and without summary), one covering unparseable dict inputs (drops silently, no crash). * test(anthropic): add non-adaptive model coverage for dict-shape reasoning_effort Per Greptile feedback on PR #28198: the original regression test only exercised the adaptive (4.6+) path. Add a parametrized test for the non-adaptive branch (claude-sonnet-4-5) verifying that dict-shape reasoning_effort still maps to thinking.type='enabled' + budget_tokens, and that output_config is NOT set on pre-4.6 models. * test(anthropic): convert unparseable-dict test to @pytest.mark.parametrize Per @greptile-apps inline review on PR #28201 — matches the parametrize style of the two adjacent dict-shape tests and produces clearer failure messages (test ID per case instead of one collapsing for-loop).

…28280) Squash-merged by litellm-agent from ro31337's PR.

oss-pr-review-agent-shin · 2026-05-19T21:16:03Z

@greptile please review

Sameerlite · 2026-05-20T10:44:59Z

https://github.com/BerriAI/litellm/pull/28310/commits has all the same commits, closing this

feat: add Xiaomi MiMo-V2.5-Pro and MiMo-V2.5 OpenRouter model entries (…

43bc7d6

…#27700) Squash-merged by litellm-agent from TorvaldUtne's PR.

oss-pr-review-agent-shin Bot mentioned this pull request May 19, 2026

feat: add Xiaomi MiMo-V2.5-Pro and MiMo-V2.5 OpenRouter model entries #27700

Merged

greptile-apps Bot reviewed May 19, 2026

View reviewed changes

fix(ui): trim whitespace from MCP inspector tool call inputs (#28203)

91927eb

Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai>

mubashir1osmani and others added 4 commits May 19, 2026 08:57

gemini-3.1-flash-lite pricing (#27933)

6f83cb2

* feat(model_prices): add gemini-3.1-flash-lite pricing with standard/batch/flex/priority tiers * fix pricing * add service tier --------- Co-authored-by: shin-berri <shin-laptop@berri.ai>

fix: incorrect /v1/agents request example (#28131)

249ec01

feat: add pricing entry for openrouter/google/gemini-3.1-flash-lite (#…

87a55f5

…28280) Squash-merged by litellm-agent from ro31337's PR.

oss-pr-review-agent-shin Bot mentioned this pull request May 19, 2026

feat: add pricing entry for openrouter/google/gemini-3.1-flash-lite #28280

Merged

5 tasks

Sameerlite closed this May 20, 2026

Sameerlite deleted the shin_agent_oss_staging_05_19_2026 branch May 22, 2026 12:07

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[litellm-agent] Staging → litellm_internal_staging (5/19/2026)#28207

[litellm-agent] Staging → litellm_internal_staging (5/19/2026)#28207
oss-pr-review-agent-shin[bot] wants to merge 6 commits into
litellm_internal_stagingfrom
shin_agent_oss_staging_05_19_2026

oss-pr-review-agent-shin Bot commented May 19, 2026 •

edited

Loading

Uh oh!

oss-pr-review-agent-shin Bot commented May 19, 2026

Uh oh!

greptile-apps Bot commented May 19, 2026 •

edited

Loading

Important Files Changed

Uh oh!

greptile-apps Bot May 19, 2026

Uh oh!

codecov Bot commented May 19, 2026 •

edited

Loading

Uh oh!

CLAassistant commented May 19, 2026 •

edited

Loading

Uh oh!

oss-pr-review-agent-shin Bot commented May 19, 2026

Uh oh!

Sameerlite commented May 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Uh oh!

Conversation

oss-pr-review-agent-shin Bot commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merged PRs (1)

Uh oh!

oss-pr-review-agent-shin Bot commented May 19, 2026

Uh oh!

greptile-apps Bot commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 5/5

Important Files Changed

Uh oh!

greptile-apps Bot May 19, 2026

Choose a reason for hiding this comment

Uh oh!

codecov Bot commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

CLAassistant commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oss-pr-review-agent-shin Bot commented May 19, 2026

Uh oh!

Sameerlite commented May 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

oss-pr-review-agent-shin Bot commented May 19, 2026 •

edited

Loading

greptile-apps Bot commented May 19, 2026 •

edited

Loading

codecov Bot commented May 19, 2026 •

edited

Loading

CLAassistant commented May 19, 2026 •

edited

Loading