LoRA is generated entirely from anime images.
Should work well with any model.
Check samples for prompts :)
Leave a like, comment, review but first of all, ENJOY ;)
If you wish to support me in my work,
you can do that here: Ko-Fi.com/LittleJelly
Description
FAQ
Comments (14)
can you make character from slow loop?
Sorry I've never seen that anime. Therefore I won't make the characters.
@LittleJelly thats anime was one universe with Bocchi The Rock
The clothing and shoe features don't seem to be restored well :sad:
Wut? Did you check the samples?
I don't have any issues with clothing at all.
For shoes try: 'white boots'
Note: There are only 3-4 scenes in the whole anime (and the dataset) where we can see her full body with boots. So I don't think that's enough for AI to reproduce it properly.
How come your LoRA models produce good hands compared to other creator LoRAs?
Talent ✌️
Naaah... To be honest it's completely random 😄
When I try to do open hands generations I tend to generate 60 images and pick the one with best hands. But out of those 60 images just 2-3 max have decent hands. The ai just doesn't know how to do proper hands yet.
@LittleJelly I only get consistent hands with your LoRA models and another creator named JustNTP
@forgerloid Magic I guess.
@LittleJelly Yep. Probably just got lucky.
Are you going to do more Mushoku Tensei characters like Hilda and Elinalise?
Your lora models look absolutely amazing, do you have any plans on making a training guide?
It's all about quality screencaps. Training guide ... I do nothing special
很棒的模型
Thx