Deerling LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. My first lora trained on a predominantly feral (70%) dataset, with very promising results. Works very consistently, you may need to add (head flower) to your prompt to get the correct flower on Deerling's head. Only does the spring form of Deerling at the moment, I plan to retrain this model once I learn how to incorporate multiple concepts into a LoRA correctly and add support for the other forms. For now, spring Deerling had the most images to train on, so it's what I decided to pick. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on an all female dataset.