Trained on 3d movies from Disney, Pixar, Dreamworks, etc. I've always wanted to make this type of model, but ended up being on the bottom of my TODO list for very long.
I think it's now or never.
Purpose of this model Make amazing 3d Toon style artworks on its own.
Train character loras where the dataset is mostly made of 3d movie screencaps, allowing less style transfer and less overfitting.
It's also pretty good at generating NSFW stuff.
Works very well with all the loras and TIs in my ecosystem, and with every well done character and concept loras that are not style overfitted.
Clone with HTTP
git clone https://www.modelscope.cn/YorickHe/3d_Animation_Diffusion.git
评论