Apple, considered by many to be very conservative in its approach to AI, has quietly released frameworks and model libraries designed to run on its chips, potentially bringing generative AI applications to MacBooks.
The company’s machine learning research team released MLX, a machine learning framework that allows developers to build models that run efficiently on Apple silicon, and MLX Data, a deep learning model library. Both are available through open source repositories such as GitHub and PyPI.
According to Apple on GitHub, frameworks like PyTorch, Jax and ArrayFire inspired the design of MLX, with the notable difference that it has shared memory, meaning any task run on MLX will work on supported devices (currently CPUs and GPUs) without moving data. Computerworld reported that MLX is designed to be easy for developers to use, but is powerful enough to train AI models such as Meta’s Llama and Stable Diffusion. Frameworks and model libraries help power many of the AI applications on the market today.
Awni Hannun, a machine learning researcher at 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 contacted Apple for more information.
However, these focused on machine learning, rather than the popular generative AI applications that competitors like Microsoft and Google have been chasing. Apple even avoids using the word AI in many of its keynote presentations.
In September, Apple reportedly began working on basic models to see which could be implemented across its services.
