BerriAI/litellm v1.74.7-nightly

on GitHub

What's Changed

• Copy MCP Server name by @NANDINI-star in #12760
• Fix Y-axis labels overlap on Spend per Tag by @NANDINI-star in #12754
• [Feat] Backend - Add support for disabling callbacks in request body by @ishaan-jaff in #12762
• Bulk Edit Users on UI by @krrishdholakia in #12763
• Litellm gemini grounding metadata stream by @krrishdholakia in #12673
• [Docs] 1.74.6.rc note by @ishaan-jaff in #12765
• feat(proxy_server.py): add model hub to the swagger by @krrishdholakia in #12767
• fix: correct Groq model naming convention for moonshotai/kimi-k2-instruct by @colesmcintosh in #12768
• [JSON Logs] fix ciruclar ref error by adding safe dumps by @jugaldb in #12764
• UI - Support 'batch' model health checks + make 'team-only' model concept clearer by @krrishdholakia in #12770
• [Key Access] Litellm disabled callbacks for UI by @jugaldb in #12769
• Fix moonshot/kimi-thinking-preview tool choice support by @colesmcintosh in #12772
• [QA] Disable Logging settings for Keys by @ishaan-jaff in #12774
• Allow forwarding clientside headers by model group by @krrishdholakia in #12753
• fix(proxy): Fix Model Armor project_id initialization order by @colesmcintosh in #12766
• [LLM Translation] added switchpoint router by @jugaldb in #12777
• [LLM Translation] Add Gov Cloud bedrock model pricing and context windows by @jugaldb in #12773
• Litellm fix proxy unit testing by @krrishdholakia in #12778
• UI - support adding links to model hub by @krrishdholakia in #12776
• Litellm fix proxy unit testing p2 by @krrishdholakia in #12779

Full Changelog: v1.74.6-nightly...v1.74.7-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.74.7-nightly

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