Alolan Vulpix LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Works incredibly well, primary issue is tail placement. Will sometimes not include the tails, so you may need to prompt for them. I got good results using (fanned tails), but (tails) should work just fine too. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on a mixed male/female dataset.