Verified: 7 months ago
SafeTensor
You reviewed this on Jul 7
Type | |
Stats | 1,000 1,334 |
Reviews | (157) |
Published | Feb 17, 2024 |
Base Model | |
Training | Steps: 4,500 Epochs: 5 |
Usage Tips | Clip Skip: 2 |
Hash | AutoV2 B98AD8A11D |
This lora was trained on about 100 images from the game series Etrian Odyssey. The lora managed to capture the style quite well. Nonetheless, it has problems with accurate hands, this is due to the data set having awkward hands.
This lora was trained on Autism Pony. I haven't tested it with anything else.
To get the results you see in the images above, these are the settings I used.
Prompt: score_9, [your prompt] <lora:Etrian_Odyssey_XL:0.8>
Negatives: mosaic censoring,bar censor,censored,source_cartoon,source_pony,source_furry,
Sample: Euler a
Steps: 33 (Whatever gives you the best results)
CFG Scale: 7
Highres Fix: 4x-UltraSharp + ADetailer