alea31415/

aohina-yamanosusume

license: creativeml-openrail-m

1. aohina yuri kurauehinata yukimuraaoi 2girls
2. yukimuraaoi girl
3. kurauehinata girl
4. aobakokona girl

• Trained with TheLastBen's fast-DreamBooth notebook
• data set: 198 concept images, I cannot remember whether I used reg images or not (it seems not)
• learning rate 1e-6 for 15500 steps
• text encoder rate 10%

• As always, it is hard to get interactions between multiple characters even though I explicitly trained for it.
• For some mysterious reasono, when generating images for aohina it can hardly get the eye colors right (maybe should train at clip_skip=2?).