So, I made a little mistake - I was trying to recreate thickness of a cum from some of my favorite checkpoints into LoRA, so I could get thickness in other checkpoints that lacks in the topic. But the data set I created to learn a LoRA had not enough diversity, so now, using it on: <code>strength: 0.9</code> and with trigger tag <code>cum on body</code>, we can get cum everywhere, on the floor, on the walls, etc. Yes, it is a broken LoRA, becouse it fails it purpose. Like - overkilling it. But maybe this is something you are looking for. I just don't want to delete it pernamently 😜But don't worry, I'll try to make another one with better images in training process, with more diversity, so the LoRA could understand the difference between "on body", "on floor" and "everywhere" 😄

Files with SHA256: 4615734a9e70a5844e9853514a29525a16a27cb17feadca1b50c9e52fcd12fdc