5 ms·
Without k-diffusion support, I don't think this replicates Stable Diffusion experience: https://github.com/crowsonkb/k-diffusion https://github.com/crowsonkb/k
by code51 4y ago
Without k-diffusion support, I don't think this replicates Stable Diffusion experience:
https://github.com/crowsonkb/k-diffusion https://github.com/crowsonkb/k-diffusion
Yes, running on M1/M2 (MPS device) was possible with modifications. img2img and inpainting also works.
However you'll run into problems when you want k-diffusion sampling or textual inversion support.
- Birch-san 4y agostable-diffusion supports k-diffusion just fine on M1. You just have to detach a tensor in to_d() to stop the values exploding to infinity. https://twitter.com/Birchlabs/status/1563622002581184517?s=20&t=kt9noe6QjD2WLvcQGKbnww https://twitter.com/Birchlabs/status/1563622002581184517?s=2...
- code51 4y agoI've been following your MPS branch and have run it but couldn't address the issue without this explanation. Thank you!