If you like the LoRa, please like it and post your image result below... :D
I only train models for characters that I or friends of mine are interested in, so don't ask if I could train character XYZ...
Took ages :<
About the LoRa
It's a character LoRa for Pony Diffusion v6 based checkpoints (I still don't understand why I should train with other pony based models...). I used 84 cleaned up images (Only from the Frozen movies and short movies) for training... It Was really 'fun' to remove any other character from each training image... I choose roundabout 10 images per outfit from the movie, sadly some of them didn't stick correctly :/ I will try to fix that in a future version (and maybe even add the stuff from wreck it Ralph ^^)...
How to use it?
Simply copy the trigger words. The base trigger word is anna. Then simply add one of the outfit combinations in the trigger list, each one is a complete outfit from the movies. I would also suggest using Adetailer to improve the face. And use 3d in positive prompt to get a style that fits the movies, or put it in negative prompt so it looks more like a drawn image. I didn't use upscale fixing at all (basically because its currently painfully slow :<)