Galarian Rapidash LoRA trained on NovelAI with Kohya training scripts using the DAdaptAdam optimizer at 512x512 resolution. Works very consistently. Should have wide compatibility with model mixes based on NovelAI, which is most of them. Trained on a mixed male/female dataset.
Description
Initial release
FAQ
Comments (3)
When I saw you posted the regular one, I was hoping there'd be the more fabulous version coming too, very nice!
No way I could leave out the fabulous Galarian forms! I made sure all of the Ponyta and Rapidash forms were ready last night so I could release them all at once today 😎