MLXCopy wiki link

MLX is an open-source array framework for machine learning on Apple silicon, developed by Apple machine-learning researchers. Its design uses the unified memory architecture of Apple silicon and supports model training and inference through Python and lower-level APIs.

MLX is a framework rather than a complete private-assistant product. Model acquisition, chat interfaces, retrieval, tool use, updates, and storage are supplied by surrounding projects and deployment choices.

Private AI competitors includes MLX because it makes capable device-local workflows practical on Macs, within the broader trust boundary described in Local AI.