Alolan Ninetales LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Very consistent, biggest issue I ran into was tail placement and count. Adding (multi tail) to my prompt when Ninetales was clothed helped me get more consistent tails. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on an all female dataset.