Repo

frink

@antonellofby Antonello Fratepietro · Independent

0

Pure-Rust GGUF inference engine with quantized CPU, Metal & CUDA kernels, MoE support, OpenAI-compatible server. Benchmarked head-to-head against llama.cpp.

Built withRustaiinferenceinference-enginellm-inferencelocal-llm

See what other members are building and what each project is looking for.