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