Verified: a year ago
SafeTensor
Type | |
Stats | 307 440 9 |
Reviews | (85) |
Published | Feb 21, 2024 |
Base Model | |
Training | Steps: 2,100 Epochs: 6 |
Usage Tips | Strength: 1 |
Hash | AutoV2 BDA2E16AFC |
Many children's book illustrators paint collages. This results in a simple, but highly stylized image. This LoRA produces images that mimic collage paintings.
Strength ~1 is usually sufficient, but you may need to go higher depending on the base model and the subject matter. I've found that sometimes it's better to do the initial generation at strength = 1.0 and then use a higher strength (1.5 to 2.0) for the upscale. Again, it all depends on the base model and the subject matter.