Litellm health check tokens

[Harshit28j]

Relevant issues

Address health check token overconsumption by introducing a configurable limit and a sensible default for non-wildcard models.

Pre-Submission checklist

• I have Added testing in the tests/litellm/ directory (Added tests/test_litellm/proxy/test_health_check_max_tokens.py)
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem
• I have requested a Greptile review by commenting @greptileai

CI (LiteLLM team)

• Branch creation CI run
  Link:
• CI run for the last commit
  Link:
• Merge / cherry-pick CI run
  Links:

Type

🆕 New Feature

Changes

• Optimization: Updated litellm/proxy/health_check.py to default max_tokens: 1 for standard health checks. This prevents health checks (like Azure OpenAI) from generating long responses, saving cost and reducing latency.
• New Config Setting: Introduced health_check_max_tokens in model_info. Users can now explicitly set the token limit for health checks in their config.yaml.
• Wildcard Alignment: Updated litellm/litellm_core_utils/health_check_helpers.py to respect the configurable token limit while maintaining a safe default of 10 for wildcard-route models.
• Testing: Added unit tests in tests/test_litellm/proxy/test_health_check_max_tokens.py covering default behavior, custom overrides, and wildcard routing safety.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Building	Preview, Comment	Feb 27, 2026 6:10pm

[CLAassistant]

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.
1 out of 2 committers have signed the CLA.

✅ Harshit28j
❌ github-actions[bot]
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[Harshit28j]

@greptile can you please review this PR?

[greptile-apps]

Greptile Summary

Introduces optimization to reduce health check token consumption by defaulting max_tokens to 1 for non-wildcard models while maintaining a safe default of 10 for wildcard routes. Adds configurable health_check_max_tokens setting in model_info for user customization.

Key Changes:

• Standard health checks now default to 1 max token, reducing cost and latency
• Wildcard models (like openai/*) maintain 10 token default to avoid errors with models like gpt-5-nano
• New health_check_max_tokens config option allows per-model customization
• Logic correctly checks both health_check_model and original model for wildcard detection
• Comprehensive test coverage with proper mocking (no real network calls)
• Poetry.lock version bump is routine maintenance

Minor Issue:

• Documentation for the new health_check_max_tokens configuration option should be added to the health.md file

Confidence Score: 5/5

• Safe to merge - well-designed optimization with proper fallbacks and test coverage
• Logic is sound with proper wildcard detection, tests are comprehensive and properly mocked, no breaking changes (user config can override defaults), only minor documentation gap
• No files require special attention - all changes are straightforward

Important Files Changed

Filename	Overview
litellm/proxy/health_check.py	Adds configurable max_tokens with sensible defaults (1 for standard models, respects wildcards)
litellm/litellm_core_utils/health_check_helpers.py	Updated to respect existing max_tokens from upstream, maintains 10 as fallback for wildcards
tests/test_litellm/proxy/test_health_check_max_tokens.py	Comprehensive tests covering default behavior, custom overrides, and wildcard safety - properly mocked

_{Last reviewed commit: 2553698}

[greptile-apps]

_{4 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

add documentation for the new health_check_max_tokens config option in /docs/my-website/docs/proxy/health.md so users know they can configure this

_{Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!}