Fastest LLM inference runtime for Apple Silicon - View it on GitHub