Verified: a year ago
SafeTensor
Type | |
Stats | 280 0 |
Reviews | (55) |
Published | Dec 30, 2023 |
Base Model | |
Training | Steps: 7,400 Epochs: 10 |
Usage Tips | Clip Skip: 2 Strength: 0.6 |
Hash | AutoV2 99FC34D37E |
Creates images in the style of artist Omar Dogan (https://www.instagram.com/omardogan1976/)
Recommended strength: 0.4-0.8
You will need to tweak the strength to find just the right amount depending on what model, LoRAs, etc you are prompting.
Resolutions of 512x512 to 1024x1024 work, but best resolution is 768x1024.
Trained on 74 scanned prints with WD14 captioning.