BerriAI/litellm v1.74.15-nightly

on GitHub

What's Changed

• Add advanced date picker to all the tabs on the usage page by @NANDINI-star in #13221
• Add Perplexity citation annotations support by @kankute-sameer in #13225
• fix: role chaining and session name with webauthentication for aws bedrock by @RichardoC in #13205
• Fix missing extra_headers support for vLLM/openai_like embeddings by @colesmcintosh in #13198
• litellm/proxy: preserve model order of /v1/models and /model_group/info by @AlexanderYastrebov in #13178
• Prompt Management - abstract prompt templates away from model list (enables permission management on prompt templates) by @krrishdholakia in #13219
• [QA Fixes for MCP] - Ensure MCPs load + don't run a health check everytime we load MCPs on UI by @ishaan-jaff in #13228
• Revert "fix: role chaining and session name with webauthentication for aws bedrock" by @krrishdholakia in #13230
• Prompt Management (2/2) - New /prompt/list endpoint + key-based access to prompt templates by @krrishdholakia in #13218
• [UI QA Fixes] Stable release by @ishaan-jaff in #13231

New Contributors

• @kankute-sameer made their first contribution in #13225
• @AlexanderYastrebov made their first contribution in #13178

Full Changelog: v1.74.14-nightly...v1.74.15-nightly

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.74.15-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