Type | |
Stats | 377 1,872 |
Reviews | (80) |
Published | Oct 24, 2023 |
Base Model | |
Trigger Words | Jalter |
Hash | AutoV2 99FBCE067E |
Jeanne d'Arc Alter (Shinjuku Ver.) LoRA
LoRA request queue on my discord: https://discord.gg/DAfjhRfyEM
This model was trained using 27 images of her. It was trained on the NAI model so it should work with most anime based models.
Recommended weight is between 0.7 and 1, I suggest going for 0.8
The triggerword is Jalter. Big disclaimer here that this one is not that great. I tried around a LOT and couldn't get the jacket right. I think SD 1.5 just isn't that great with jackets that have different colors on the outside and inside. So expect to see color bleed. I also suggest adding holding, holding sword in the negative prompt.
Example images were generated using jafamix v4 with 30 steps, then hires. fixed by 1.5 with adetailer enabled for the faces
Note: This was made exclusively on anime images and on an anime checkpoint. Feel free to post images on realistic, semirealistic etc. checkpoints, but do not rate it on that! This was made to be used on anime checkpoints so keep reviews for that!