Trained on 500 low quality images with hand fixed captions (still bad). Dataset was originally intended to train sd 1.5, but I've never got to actually train it. The quality of this lora is better than I expected but still kind of bad, so don't expect anything.For some reason only work with strength of 1.3. Use word crlty as activation. Also model haven't learned concept like Foxtrot, Handler, Garbino, etcPlease, donate buzz if you want to see anything else Cruelty Squad related (actually good this time)

Files with SHA256: f0e2baaae8b773f617d05a7f235df9ea79bdc7fee5b240cee0360a6df55683f3