This is a Stable Diffusion 2.1 768 model trained on many images of disco diffusion in an attempt to emulate the disco diffusion style.

I am happy with the outputs which are varied and should give the user some of those wild disco diffusion type images.

If you have an updated version of automatic1111 you should not need a custom yaml, as I did not need. If you do not have a updated version for whatever reason rename the checkpoint yaml file copy to the same name as this custom checkpoint.

Trigger word for this model is discostyle.

Share some images if you like the model.