Apple’s machine learning research team introduced a machine learning framework called MLX. In this way, developers can create efficient models and deep learning model libraries for Apple Silicon. Both the framework and MLX Data-based deep learning models are available on the GitHub page and via PyPI.
Apple is also preparing to advance in the field of artificial intelligence
According to Apple’s GitHub page, MLX, inspired by PyTorch, Jax and ArrayFire, will have shared memory. This means that any task running on MLX will be able to run on all supported hardware without the need to move data. Currently CPUs and GPUs have MLX support. According to reports, MLX is simple enough for developers to use, but has enough power to develop models such as Meta’s Llama and Stable Diffusion generative AIs. Frameworks and model libraries can be used for many AI applications on the market.
Still, MLX Data focuses on deep learning in general. Apple does not focus on the field of productive artificial intelligence like its major competitors Microsoft and Google. In fact, the company avoids using the word artificial intelligence as much as possible.
It was previously stated that Apple was working on basic models. The underlying models contain a lot of unlabeled data. Thus, it is possible to use the same data set for different purposes.
53889
