Skip to content

Comment on Stable Diffusion launch announcementparent

Comments

Right now the model people use in the closed beta is really permissive, it can create NSFW, famous people, gorish images, etc. I don't think (I hope) that the model will change much. They will probably release all the weights from the different training phases.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.