Key Info

LiteLLM's weekly update includes a Customer Portal beta for attributing traffic and inference spend to individual customers, teams and members, a fully Rust-based /v1/messages endpoint, Day 0 support for several new Claude and Gemini models across Anthropic, Vertex AI, Bedrock, Azure and GitHub Copilot, and Replicate integration for video, image and speech models.

Highlights

  • Customer Portal (beta): serve your own customers through LiteLLM and attribute traffic and inference spend per customer, team and member.
  • /v1/messages is now fully implemented in Rust.
  • Day 0 model support: Claude Fable 5.2, Claude Mythos 5.2, Gemini 4 Carbon, Gemini 4 Hydrogen and Gemini Decision 1 across Anthropic, Vertex AI, Bedrock, Azure and GitHub Copilot.
  • Replicate integration adds video, image and speech models via /v1/videos, /v1/images/generations and /v1/audio/speech.
  • Weekly office hours are held live on Fridays at noon Pacific.