[https://nvbugs/6114711][fix] Add "kimi_k2": "kimi_k2" and "kimi_k25": "kimi_k2" to `MODEL_TYPE_TO_REASONI

[tensorrt-cicd]

Summary

• Root cause: MODEL_TYPE_TO_REASONING_PARSER dict was missing entries for kimi_k2 and kimi_k25 model types, so resolve_auto_reasoning_parser() returned None and no reasoning parser was applied, causing <think>...</think> content to be returned verbatim in the response.
• Fix: Add "kimi_k2": "kimi_k2" and "kimi_k25": "kimi_k2" to MODEL_TYPE_TO_REASONING_PARSER dict, and update the error message in serve.py to list these model types.
• Automated fix generated by repair-bot

Test plan

• Verify fix on the same GPU type as the original failure
• Check for regressions in related tests

Links

• Bug: https://nvbugs/6114711

Summary by CodeRabbit

• New Features
  • Added support for kimi_k2 and kimi_k25 model types in reasoning parser auto-detection functionality.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 45dcfa38-729f-4dd6-8e48-680ffff06a96

📥 Commits

Reviewing files that changed from the base of the PR and between 2f745de and 8231b61.

📒 Files selected for processing (2)

• tensorrt_llm/commands/serve.py
• tensorrt_llm/llmapi/reasoning_parser.py

---

📝 Walkthrough

Walkthrough

The changes add support for two additional model types, kimi_k2 and kimi_k25, in the reasoning parser system. A mapping is added to recognize these model types, with kimi_k25 routed to the kimi_k2 reasoning parser. The auto-detection error message is updated to reflect these newly supported types.

Changes

Cohort / File(s)	Summary
Reasoning Parser Support tensorrt_llm/llmapi/reasoning_parser.py, tensorrt_llm/commands/serve.py	Added kimi_k2 and kimi_k25 model types to the reasoning parser auto-resolution mapping, with kimi_k25 mapped to kimi_k2. Updated the error message for failed auto-detection to include these new model types in the list of supported options.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly indicates a fix to add specific model types to the MODEL_TYPE_TO_REASONING_PARSER mapping, directly corresponding to the actual changes made.
Description check	✅ Passed	The description clearly explains the root cause, the fix applied, and includes test verification and relevant bug link, addressing the core requirements of the template.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}