This is a model that's based on the style of magazine cover it will make random text and shapes in front of and behind the character without being too overpowering. Tends to work best at 0.6. This model was trained on NAI anime-full-pruned and generated in AOM3. Made with the Kohya tool for AI Training. Any suggestions to improve or recommended things for the data set would be appreciated.
Description
Works good at the moment I haven't seen any errors, and I haven't had an occurrence that required more trigger words. Works with other characters and Lora fine.
FAQ
Comments (3)
Did you train with images of multiple characters on the same cover? What about male characters?
Set of 26 pics 5 were 2 characters, 1 was 3 characters, 3 were close-ups the rest 1girl, no males theoretically should be able to do it with a character LoRA, I don't have any to test though. I'll note for an update to try to at least throw a couple males I bet there's a genshin cover somewhere if i look hard enough.
@rulles Great, thanks for the info and thanks in advance for adding in some males next time. :) I'll have to try it with a Genshin LoRA. I know there's a Genshin "splash" LoRA, but I've not seen any with covers... but I don't claim to be the best informed person.