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