Stable Diffusion with Core ML on Apple Silicon

Atila from Apple on the expected performance:

> For distilled StableDiffusion 2 which requires 1 to 4 iterations instead of 50, the same M2 device should generate an image in <<1 second

Holy hell what the fuck convergence in less than 5 iters?! How did they manage that!

>macOS 13.1 and iOS 16.2

I would have expected iPadOS, this technology is much more suited for the iPad than the iPhone.

If the author reads the comments here, the GitHub link at the end of the article is broken.