For a request.
Use between weight 0.5 and 1. Keyword is puccilavmin. Her hair ornaments are prompted with rabbit_hair_ornament and pink_bow. Standard outfit was prompted with blue_dress, blue_skirt, red_thighhighs, and boots. Optional is headphones_around_neck.
Issues
Colors are washed out, likely due to original training images having very low contrast and low saturation.
rabbit_hair_ornament is near impossible to get right. Add manually.
Pic 1: AAM - AnyLoRA Anime Mix - Anime Screencap Style Model
Pics 2-4: AnyLoRA
Description
Trained on 49 images.
FAQ
Comments (1)
Wow, the turnaround on this was incredibly fast, and the results are really good considering what was available to work with! Thanks @jwp!
I noticed that this was trained with 49 images. What did you end up using, aside from booru sites? Maybe more saturation and contrast would theoretically be possible as more art trickles in over the months...
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.