Yveltal LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Body markings are inconsistent. Body should be primarily red with black accents. Hands work with embeddings that enforce a human hand shape. 'Horn' shape can be inconsistent. Tail comes out skinnier than it should sometimes, may require adding (thick tail) to your prompt. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on an all female dataset.