Verified: 5 months ago
SafeTensor
Type | |
Stats | 221 |
Reviews | (26) |
Published | Jun 21, 2024 |
Base Model | |
Training | Steps: 2,000 Epochs: 10 |
Usage Tips | Clip Skip: 2 |
Hash | AutoV2 4CBEA8F391 |
LoRA created from 100 images with art by 武田弘光:
Recommended weight 0.7 to 1.0 depending on your model and prompt.
No trigger word is needed, but you can use 'takeda_hiromitsu' if you want.
A good style. Since trained half on monochromatic manga images and half on screenshots of a couple games, will have slight bias towards some hair colors and styles.