Scolipede LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Works shockingly well. Main issue is that it'll give her only one tail instead of the proper two. Adding (two tails) to your prompt will help get the proper tail count. Most other details are pretty consistent. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on an all female dataset.