Looks interesting, but also somewhat complicated. Now generation has to run through 3 models? I thought it was bad enough having a separate refiner model for sdxl.
this post was submitted on 13 Feb 2024
28 points (100.0% liked)
Stable Diffusion
4310 readers
5 users here now
Discuss matters related to our favourite AI Art generation technology
Also see
Other communities
founded 1 year ago
MODERATORS
It's based on Würstchen which has real savings on training, but the VRAM requirement is through the roof.
Sdxl was 3 models too. Base, refiner and vae.
Cascade is 3. Lores, hires, vae.
Oh wow, I was just looking for updates on this yesterday and suddenly this pops up! Very exciting!