Is it running through, eg, TensorFlow or PyTorch (both of which are in a constant state of better use of SoC hardware), or via custom Metal Performance Shaders, or via use of Apple's API's, or what?

This time, rahulunair uses a 16GB model of Intel Arc A770 to run Stable Diffusion. To run Stable Diffusion on Intel Arc, you need the machine learning libraries PyTorch or TensorFlow .

PyTorch was developed by the artificial intelligence research team of Facebook (currently Meta), and at the time of writing the article, it is also used in Stable Diffusion, an image generation AI.

That means developers will soon be able to run MLX models directly on NVIDIA GPUs, which is a pretty big deal. Here’s why.