No real human cosplay or fanart was used in the training of this LoRA, except for my own.
While testing variations in training data for my main troll LoRA, I made a smaller one of just Sollux, since I had the best, most representative training data for him. Since it may be a while before I'm done with the big LoRA, here is the Sollux one!
Description
The only thing you should need to specify about his outfit is his glasses, because the LoRA is configured to be able to generate images of him with or without them. You should not need to specify anything about his horns, skin, etc - that should all work automatically. I have had good results with a strength of 0.8.