This LoRa try to mimic the simple illustration style from kids book.
!!!!!!!!! PLEASE DON'T POST LEWD IMAGES IN GALLERY, THIS IS A LORA FOR KIDS ILLUSTRATION !!!!!!!!!
Training LoRa necessitate a lot of coffee. If you think that what I do is useful and want to help me keep doing this you can buy me a coffe here: https://ko-fi.com/clumsy_trainer
Description
First version, already working on a second version
FAQ
Comments (10)
what are all those negatives in your samples, ngtv?
One is the negative embedding from realistic vision, the others I can't remember since I've changed their default names. The don't change the output so much, I use them especially for hands.
Maybe I'm missing something but I cant get this to run. Am I missing a .yaml file somewhere?
no, it's a normal LoRa model, you use it with any 1.5 based checkpoint and it should work
@signorlimone Ahhh! Figured it out. Thanks! Works great!
Where is COOLKIDS_V2? I smart download Coolkids
what?
@signorlimone
I encounter a difficulty, and the same parameters cannot generate the same picture
me：
Original address：
this is really strange, are you using V1 or V2?
@Clumsy_Trainer According to their image EXIF data they are using V1 and that's why they didn't get the same result.