Ollama is now powered by MLX on Apple Silicon in preview
Summary
Ollama previews MLX-based acceleration on Apple Silicon, delivering faster time-to-first-token and tokens-per-second for local AI tasks. The update includes NVFP4-enabled precision, cache enhancements, and clearer guidance for getting started, positioning Ollama as a higher-performance option for macOS developer workflows. Acknowledgments point to collaboration with MLX, NVIDIA, and other partners.