This model is quite finicky, so if you want similar (or maybe better) results to the ones in my examples, check the images attached to the post for more information on their generation methods. The model was made with OneTrainer, with ~200 images running 20 epochs twice, but for some reason it's quite flexible, and can be used in lower strengths to mix with other people models.There's a &gt;very tiny&lt; chance this LoRA creates black and white images when not prompted. I have manually checked and corrected the source images to ensure there were no black and white or sepia images, but it's a rare occurrence and I'm investigating why this happens.

Files with SHA256: dfafe0290e343976f3be94da185b6b4af48982b00e9e7871ed23e7ae6c38e392