Michele LaurelliBlog

MLX

MLX: An Apple numerical framework for machine learning with arrays and graphs optimized for Apple silicon. It uses unified memory and lazy evaluation for training and inference on Apple devices. The MLX-LM ecosystem supports language models and specific conversion workflows.

In italiano: MLXLegge IV · Loss, gradiente e ottimizzatori

It uses unified memory and lazy evaluation for training and inference on Apple devices. The MLX-LM ecosystem supports language models and specific conversion workflows.

Examples

  • LLMs on Mac
  • Lazy array operations
  • Fine-tuning on Apple silicon

Further reading

Missing but Necessary Software Infrastructure for AI on Apple Silicon: the Triton Language