Apple drops new MLX machine learning framework for Apple silicon Macs

Don’t ask me what any of this means, but it might be of interest for some of you real Mac users. Apple has released MLX, “an array framework for machine learning on Apple silicon, brought to you by Apple machine learning research.”

This story is supported by Mosyle, the only Apple Unified Platform. Mosyle is the only solution that fully integrates five different applications on a single Apple-only platform, allowing businesses and schools to easily and automatically deploy, manage, and protect all their Apple devices. Over 38,000 organizations leverage Mosyle solutions to automate the deployment, management, and security of millions of Apple devices daily. Request a FREE account today and discover how you can put your Apple fleet on auto-pilot at a price point that is hard to believe.

Here’s a description from the documentation:

MLX is designed by machine learning researchers for machine learning researchers. The framework is intended to be user-friendly, but still efficient to train and deploy models. The design of the framework itself is also conceptually simple. We intend to make it easy for researchers to extend and improve MLX with the goal of quickly exploring new ideas.

They also highlight some key MLX features:

• Familiar APIs: MLX has a Python API that closely follows NumPy. MLX also has a fully featured C++ API, which closely mirrors the Python API. MLX has higher-level packages like mlx.nn and mlx.optimizers with APIs that closely follow PyTorch to simplify building more complex models.
• Composable function transformations: MLX has composable function transformations for automatic differentiation, automatic vectorization, and computation graph optimization.
• Lazy computation: Computations in MLX are lazy. Arrays are only materialized when needed.
• Dynamic graph construction: Computation graphs in MLX are built dynamically. Changing the shapes of function arguments does not trigger slow compilations, and debugging is simple and intuitive.
• Multi-device: Operations can run on any of the supported devices (currently, the CPU and GPU).
• Unified memory: A notable difference from MLX and other frameworks is the unified memory model. Arrays in MLX live in shared memory. Operations on MLX arrays can be performed on any of the supported device types without moving data.

Now, a thread from someone who actually knows what they’re talking about:

Enjoy!