Moritz

Here is new research for finetuning diffusion models that includes scripts for stable diffusion https://github.com/rinongal/textual_inversion/tree/main/configs/stable-diffusion

Thanks for your hard work you put into the project! I was wondering whether there is any chance we'll see a Material You implementation because there currently isn't any subsonic...

That step can take quite some time without feedback. Disable your antivirus and firewall during installation.

The card is older and it has a different design. Newer cards have stuff like tensor cores and half precision float support, that's why they aren't identical I assume.

Someone already implemented it apparently https://github.com/Birch-san/stable-diffusion/compare/34556bc45211e0a1d3554ee0cd6795793889fbfb...1f13594371a31c417c8ce5f296fead63e43be337 https://twitter.com/Birchlabs/status/1578141960249876482

Codeberg should be a good mirror, it has the best UI imo and the devs are very 'libre' oriented

Add `--no-half-vae` to your command line arguments in webui-user.bat

You have too little VRAM. Use smaller resolutions or the optimized fork https://github.com/basujindal/stable-diffusion

Someone did a manual translation in a newer PR, I think that's the way going forward

Make the traffic from python, git and conda bypass your proxy