Quiet medium quality LoRA I would say, but that's my first experience. I also trained it with low amount of images. But the facial part should be great.There are some bugs of displaying book badge she has on her shoulder and some parts of clothes would be buggy as well.

Paige Deadlock

❗This version is quite buggy cause of the small dataset.This is my first time I've made a LoRA, so don't hit me please.Usually 'paige' and 'rounded glasses' are enough, but the more trigger words you use - the more accurate it is.Also there are different sides: front, side, from behind, top, bottom. But I haven't tested them much so that's might be buggy as well.Sentences you might want to try:<ul><li>Her hair is pulled back in a neat bun</li><li>Holding a stack of books in her hand</li></ul>

Files with SHA256: 0e606357dd51f116d1917bb18452efe3deab52017ad0ff6fa66affbbdf1fc7cf