🎞 LoRA is trained purely on anime screencaps from Season 1
🗝 Activation tags are in the right panel
📝 Note: Use with 'aged up' if she generates looking too young. Season 1 had a bit weird chara scaling. Her hair ribbon is for some reason random and refuses to generate all the parts. So pray to the RNG god to get all the parts.
📸 Always use ADetailer to get detailed eyes
🛠 Technical notes:
LoRA shoud be compatible with Illu models & NoobAI. (Not tested.)
LoRA is compatible with various styles - Lower the weight & remove "anime screencap" from the prompt, if you are getting too strong anime style
Hi. res fix used on all sample images:
Upscaler: R-ESRGAN 4x+ Anime6B
Steps: 10-15
Denoising: 0.3
📬Leave a like, comment, review but first of all, ENJOY ;)
Description
FAQ
Comments (5)
Your lora is damn good, but why are you avoiding the lora like "all in one" as a pack???
Simple answer: Generation stability.
@LittleJelly It's too bad, but it's so convenient.
@GAOLD Not really convenient if you need to generate more images just to get the chara right. :)
@LittleJelly I do generally prefer quality over convinence.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.