Change8

v1.82.1-silent-dev2

📦 litellmView on GitHub →
2 features🐛 14 fixes🔧 9 symbols

Summary

This release focuses heavily on bug fixes across various integrations, including completion bridges, routing, proxy endpoints, caching, and specific provider integrations like Vertex AI, Fireworks, Sagemaker, and Bedrock. New features include silent metric recording and helmchart deployment strategy support.

✨ New Features

  • Added strategy configuration to deployment for helmchart.
  • Implemented recording of silent metrics.

🐛 Bug Fixes

  • Fixed handling of ResponseApplyPatchToolCall in the completion bridge.
  • Router now breaks the retry loop on non-retryable errors.
  • Fixed invalid OpenAPI schema for /spend/calculate and /credentials endpoints in the proxy.
  • Preserved usage/cached_tokens in Responses API streaming bridge.
  • Injected default_in_memory_ttl in DualCache async_set_cache and async_set_cache_pipeline.
  • Applied server root path to mapped passthrough route matching.
  • Merged parallel function_call items into a single assist response for the Responses API.
  • Handled month overflow in duration_in_seconds calculation for multi-month durations.
  • Used correct divisor when averaging TTFT in lowest-latency routing.
  • Stripped duplicate /v1 from the models endpoint URL for Fireworks.
  • Added role assumption support for Sagemaker embedding endpoints.
  • Stripped LiteLLM-internal keys from extra_body before merging to Gemini request for Vertex AI.
  • Preserved the reasoning_effort summary field for the Responses API for OpenAI.
  • Populated completion_tokens_details in the Responses API for Bedrock.

Affected Symbols