Solgaleo LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. A lot more consistent than I was expecting. I had to add (thin tail) to my prompt to get the tail to consistently appear. Tail detail is inconsistent. If you want cool looking eyes on YiffyMix, add (digital eyes) to your prompt. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on a predominantly male dataset.