The definitive weekly briefing on inference.

Every commit, PR, release, and issue across the open-source AI inference ecosystem — cloud serving engines, local runtimes, and edge frameworks. Delivered weekly.

Free. Unsubscribe anytime.

Tracked Repositories

View all →

127 open-source inference repositories across 59 organizations — from vLLM and SGLang in the cloud to llama.cpp and Ollama locally to ExecuTorch and LiteRT on mobile/edge.