Running local models on Macs gets faster with Ollama's MLX supportarstechnica.com2 points·by kinoriw·3 mesi fa·0 comments