Native Rust/C++ LLM inference engine

A native Rust/C++ LLM inference engine delivers high performance and low latency, making it suitable for prototypes and internal workflows. Adoption involves ma