Based on the works of nnff

This LoRA was trained using pre-2023 images, so its output may not closely resemble nnff's more recent posts.

I'll probably train a separate model for the more modern dataset.