trufty

I ended up training Unet1 separately from Unet2 and then splicing the models together as a workaround.

Duplicate of https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/1266

Yea the output will not be compatible since its in the Diffusers format. We either wait for a converter (which may be in the works). Or see if one of...

Yea you can't just drop in H-14. SD would need to be retrained from scratch. However Clip guided diffusion could be implemented which is a somewhat less effective (and slower)...

CLIP-Guided Diffusion basically runs clip against every step of the generation to push the next step more towards the desired direction. It should lead to better output, but I haven't...