Running local models on Macs gets faster with Ollama's MLX support

Ollama added support for Apple’s open source MLX framework to run local models.

Ollama also improved caching and added Nvidia NVFP4 model compression support.

The new features, in preview in Ollama 0.19, currently support only Qwen3.5 35B and target M1-or-later Macs with at least 32GB RAM.