Apple Silicon and macOS VMs: 11–16× Faster LLM Inference with Llama.cpp

(github.com)

197 points | by frabonacci 4 hours ago ago

22 comments