Running local models on Macs gets faster with Ollama's MLX supportarstechnica.com2 points·kinoriw··0 commentsOpen articleSaveView on HN