Last released Nov 12, 2026
Autotuner for LLM serving: measures vLLM, SGLang and llama.cpp settings on your GPU and your traffic, then serves the fastest
Last released Sep 15, 2026