BerriAI/litellm v1.43.18

on GitHub

What's Changed

• [Feat] return x-litellm-key-remaining-requests-{model}: 1, x-litellm-key-remaining-tokens-{model}: None in response headers by @ishaan-jaff in #5259
• [Feat] - Set tpm/rpm limits per Virtual Key + Model by @ishaan-jaff in #5256
• [Feat] add prometheus metric for remaining rpm/tpm limit for (model, api_ley) by @ishaan-jaff in #5257
• [Feat] read model + API key tpm/rpm limits from db by @ishaan-jaff in #5258
• Pass-through endpoints for Gemini - Google AI Studio by @krrishdholakia in #5260
• Fix incorrect message length check in cost calculator by @dhlidongming in #5219
• [PRICING] Use specific llama2 and llama3 model names in Ollama by @kiriloman in #5221
• [Feat-Proxy] set rpm/tpm limits per api key per model by @ishaan-jaff in #5261
• Fixes the tool_use indexes not being correctly mapped by @Penagwin in #5232
• [Feat-Proxy] Use model access groups for teams by @ishaan-jaff in #5263

New Contributors

• @dhlidongming made their first contribution in #5219
• @Penagwin made their first contribution in #5232

Full Changelog: v1.43.17...v1.43.18

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.43.18

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