A test lora i created to learn how to train, best results i get are at 0.75 weight. Still needs some tweaking but when it works....oh boy
Created with Kohya_ss so perhaps you will need the Kohya LORA extension in AUTOMATIC1111 to use it.
-Trained with 768x768
-Added more variation and "sharpen" her eyes
-Works better with lower weight, but still recommend (0.5-0.75)
-more tags still create some artifacts
-still quite finicky
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
