Apple, which many had considered very conservative in its approach to AI, quietly released frameworks and model libraries designed to run on its chips and maybe bring generative AI apps to MacBooks.
The company’s machine learning research team released MLX, a machine learning framework where developers can build models that run efficiently on Apple Silicon and deep learning model library MLX Data. Both are accessible through open-source repositories like GitHub and PyPI.
According to Apple on GitHub, frameworks like PyTorch, Jax, and ArrayFire inspired the design of MLX, with the notable difference of having a shared memory, meaning any task run on MLX works on supported devices (right now, CPUs and GPUs) without moving data. Computerworld reported that MLX is intended to be easy to use for developers but has enough power to train AI models like Meta’s Llama and Stable Diffusion. Frameworks and model libraries help power many of the AI apps in the market now.
Awni Hannun, a machine learning researcher with Apple, tweeted that MLX Data is a “framework agnostic, efficient, and flexible package for data loading” and works with MLX, PyTorch, or Jax frameworks. The Verge reached out to Apple for more information.
Apple has previously worked with AI, embedding the technology into its products for years
However, these focused on machine learning and not the popular generative AI applications that competitors like Microsoft and Google have been chasing. Apple even steers clear of using the word AI in many of its keynote presentations.
In September, Apple reportedly began working on foundational models to see which can be implemented across its services.