⚡ Pure-Rust WebGPU inference engine — OpenAI-API compatible, GGUF native, runs on any GPU. No Python. No llama.cpp. Single binary.
Michael-A-Kuykendall/shimmy has 5,729 GitHub stars.
It currently ranks #36008 daily, #36239 weekly, #37400 monthly by momentum on Trending Repos; its best-ever rank is #27670.