Why this repo matters
Latest release 1d ago, 9 developer signals, 3 package/install signals, +4 stars/7d
Local inference you can prove. A single-binary GGUF engine in C: serves, scores, and trains LoRA through the quantized weights it deploys, byte-reproducibly. Tool calls survive the token limit; every run replays as a signed receipt. CPU/CUDA/Metal, OpenAI- and Anthropic-compatible. (22 stars, 0 forks, C, fresh release, 4 AI signals, 4 developer signals). Latest release: v1.2.1.