LiteLLM Files

What's Changed

• (Bug) Fix reasoning response ID by @Sameerlite in https://github.com/BerriAI/litellm/pull/15265
• fix gemini cli by actually streaming the response by @Sameerlite in https://github.com/BerriAI/litellm/pull/15264
• Add gpt-realtime-mini support by @Sameerlite in https://github.com/BerriAI/litellm/pull/15283
• [Feat] Proxy CLI - dont store existing key in the URL, store it in the state param by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15290
• Fix: Make PATCH /model/{model_id}/update handle team_id consistently with POST /model/new by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15297
• [Fix] Networking: remove limitations by @AlexsanderHamir in https://github.com/BerriAI/litellm/pull/15302
• [MCP Gateway] Litellm mcp fixes team control by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15304
• [MCP Gateway] QA/Fixes - Ensure Team/Key level enforcement works for MCPs by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15305
• fix: model + endpoints page crash when config file contains router_settings.model_group_alias by @ARajan1084 in https://github.com/BerriAI/litellm/pull/15308
• Upgrades tenacity version to 8.5.0 by @ARajan1084 in https://github.com/BerriAI/litellm/pull/15303
• [QA/Fixes] - Dynamic Rate Limiter v3 - final QA by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15311
• Add Cohere Embed v4 support for AWS Bedrock by @timelfrink in https://github.com/BerriAI/litellm/pull/15298
• fix(bedrock): include cacheWriteInputTokens in prompt_tokens calculation by @timelfrink in https://github.com/BerriAI/litellm/pull/15292
• fix issue with parsing assistant messages by @Sameerlite in https://github.com/BerriAI/litellm/pull/15320
• feat(files): add @client decorator to file operations by @FelipeRodriguesGare in https://github.com/BerriAI/litellm/pull/15339
• [Fix] Watsonx - Apply correct prompt templates for openai/gpt-oss model family by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15341
• potentially fixes a UI spasm issue with an expired cookie by @ARajan1084 in https://github.com/BerriAI/litellm/pull/15309
• Add gpt-5-pro-2025-10-06 to model costs by @sandeshghanta in https://github.com/BerriAI/litellm/pull/15344
• Fix - (openrouter): move cache_control to content blocks for claude/gemini by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15345
• [Fix] x-litellm-cache-key header not being returned on cache hit by @ishaan-jaff in https://github.com/BerriAI/litellm/pull/15348
• Add native Responses API support for litellm_proxy provider by @Copilot in https://github.com/BerriAI/litellm/pull/15347
• AzureAD Default credentials - select credential type based on environment by @krrishdholakia in https://github.com/BerriAI/litellm/pull/14470
• SSO - support EntraID app roles by @krrishdholakia in https://github.com/BerriAI/litellm/pull/15351
• MCP - support converting OpenAPI specs to MCP servers by @krrishdholakia in https://github.com/BerriAI/litellm/pull/15343
• LiteLLM UI Refactor Infrastructure by @ARajan1084 in https://github.com/BerriAI/litellm/pull/15236
• MCP - specify allowed params per tool by @krrishdholakia in https://github.com/BerriAI/litellm/pull/15346

Full Changelog: https://github.com/BerriAI/litellm/compare/v1.77.7.dev.3...v1.77.7.dev9

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.77.7.dev9

Don't want to maintain your internal proxy? get in touch 🎉

Hosted Proxy Alpha: https://calendly.com/d/4mp-gd3-k5k/litellm-1-1-onboarding-chat

Load Test LiteLLM Proxy Results