BerriAI/litellm v1.72.6.dev1

on GitHub

What's Changed

• Time taken column logs by @gbrian in #11723
• UI - Fix remaining users activity if no limit + allow filtering by model access groups by @krrishdholakia in #11730
• [Feat] Prometheus - Allow controlling exposed metrics and labels by @ishaan-jaff in #11489
• Add Langfuse OpenTelemetry Integration by @colesmcintosh in #11607
• Fixed grok-3-mini to not use stop tokens by @Naggafin in #11563
• Implement health check backend API and storage functionality by @colesmcintosh in #11678
• fix: re-export get_valid_models utils in init file by @jsondai in #11721
• [Bug Fix] Ensure /redis/info works on GCP Redis by @ishaan-jaff in #11732
• [Fix] JWT - Fix error when team member already part of team by @ishaan-jaff in #11735
• [Bug Fix] Rollback dd-trace version by @ishaan-jaff in #11736
• Prometheus - fix request increment + add route tracking for streaming requests by @krrishdholakia in #11731
• Litellm stable release 06 14 2025 by @krrishdholakia in #11737

New Contributors

• @gbrian made their first contribution in #11723
• @Naggafin made their first contribution in #11563
• @jsondai made their first contribution in #11721

Full Changelog: v1.72.6-nightly...v1.72.6.dev1

Docker Run LiteLLM Proxy

docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
ghcr.io/berriai/litellm:main-v1.72.6.dev1

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