Verified: 9 months ago
SafeTensor
Type | |
Stats | 143 11 1 |
Reviews | (23) |
Published | Aug 13, 2024 |
Base Model | |
Training | Steps: 250 Epochs: 7 |
Usage Tips | Clip Skip: 2 |
Hash | AutoV2 D8524323FB |
Quan Anime Style:
First time training style may not work very well.
1024x1024, 512x512, 768x768 portrait styles
May use simple background, white background
Lower weights if hand deforms
use txt2img hi-res
Recommended Running Settings:
Highly Recommended to use SDFORGE (if lower VRAM) using hires fix to txt2img outputs
Hires.fix Upscaler: R-ESRGAN 4x+ Anime6B
Upscale 1.5-2
Hires Steps 20-25
Sampling Steps 25-40
Optional: Adetailer for face detail
Model Issues
Hands issues
may not work with some models