Lora trained on the art of virtual YouTuber Usada Pekora.
Trained on AOM2
Recommended weight 0.6-0.8
Important: you need to specify in the promte the tags of what you want, for example, carrot hair, gloves, orange eyes, pantyhose, rabbit ears, etc. Because lora can often forget to sketch out such details (yes, this is a bit inconvenient).
Unfortunately, it doesn't work very well so far. Often not beautiful backgrounds are obtained, often the picture can be cropped, perhaps the description of the background in the prompt and cropped in the negative prompt will help. So far, he can draw only one of the styles of Usada Pekora. Perhaps in the future I will redo this lora, but for now I'm too lazy.
Description
learning
Tool: Kohya's GUI
50img * 100steps / batch size 2
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.







