●Change8

litellm_1.81.13-dev

📦 litellmView on GitHub →
✨ 8 features🐛 24 fixes🔧 24 symbols

Summary

This release introduces a new Prompt Management API and significant UI enhancements, including access group selectors and usage filtering. It also includes extensive bug fixes focused heavily on improving test stability and isolation across various providers and features.

Migration Steps

  1. If you were using Reasoning() for the reasoning parameter, change it to use None instead.

✨ New Features

  • Introduced a new Prompt Management API for interacting with Prompt Management integrations.
  • Added Access Group Selector to Create and Edit Flow in the UI (Keys/Teams section).
  • Added filtering by User in the UI (Usage section).
  • Added support for Databricks in response schema.
  • Added new models: github_copilot/gpt-5.3-codex and github_copilot/claude-opus-4.6-fast.
  • Added guardrail tracing UI showing policy, detection method, and match details.
  • Added EU AI Act Article 5 policy template for prohibited practices detection.
  • Added GDPR Art. 32 EU PII Protection Policy Template.

🐛 Bug Fixes

  • Preserved key_alias and team_id metadata in /user/daily/activity/aggregated after key deletion or regeneration.
  • Replaced usage of Reasoning() with None for the reasoning parameter.
  • Fixed Langfuse test isolation to prevent flaky failures.
  • Resolved merge conflict and fixed bedrock thinking test flakiness.
  • Fixed flaky tests caused by module reloading and missing mocks.
  • Fixed SSO test flakiness by correctly mocking premium_user.
  • Made policy_resolve_endpoints importable without FastAPI.
  • Added cleanup for disable_aiohttp_transport in test_extra_body_with_fallback.
  • Updated reasoning_effort test to expect dict format.
  • Fixed test environment cleanup for Vertex AI rerank, GPT-OSS, and Qwen tests.
  • Used async side_effect for client.post mock in watsonx test.
  • Mocked vertexai module in GPT-OSS tests to prevent authentication issues.
  • Updated test_other_constraints_preserved for new schema filtering.
  • Fixed chatgpt codex unsupported params issue.
  • Fixed converse anthropic usage object according to v1/messages specs.
  • Fixed miscellaneous bugs in Managed batches.
  • Clamped thinking.budget_tokens to a minimum of 1024 for bedrock.
  • Fixed virtual key grace period retrieval from env/UI.
  • Fixed SSO PKCE support failure in multi-pod Kubernetes deployments.
  • Fixed missing DATABASE_URL handling in proxy's append_query_params.
  • Reverted StreamableHTTPSessionManager to stateless mode in MCP.
  • Fixed OCI Grok output pricing.
  • Fixed au.anthropic.claude opus 4 6 v1 issues.
  • Fixed Key Expiry Default Duration.

Affected Symbols

/user/daily/activity/aggregatedpolicy_resolve_endpointsReasoningLangfuse clientbedrock thinkingSSOtest_video_content_handler_uses_get_for_openaitest_extra_body_with_fallbacktest_openai_env_baseVertex AI rerank testsreasoning_effort testVertex AI GPT-OSS testsVertex AI Qwen testswatsonx testtest_other_constraints_preservedchatgpt codexconverse anthropic usage objectbedrock thinking.budget_tokensvirtual key grace periodSSO PKCEappend_query_paramsStreamableHTTPSessionManagerOCI Grok output pricingKey Expiry Default Duration