This model is an aesthetic fine-tune of Illustrious XL (v0.1-2.0), specialized for anime-style generation of characters, clothing, and scenes. It employs v-prediction with ZTSNR, was trained using Uniform LogSNR sampling on curated datasets, and maintains LoRA compatibility with its bases.<hr />Styles: <a target="_blank" rel="ugc" href="https://github.com/spawner1145/comfyui-lsnet">LSNet</a> along with the <a target="_blank" rel="ugc" href="https://huggingface.co/heathcliff01/Kaloscope2.0">Kaloscope</a> model can query artist tags from any imageTools: <a target="_blank" rel="ugc" href="https://github.com/geltz/comfyui-geltz">Custom node pack</a>Recommended nodes: Quantile Match Scaling, Token Sculptor, Token Shuffler, Perturbed Attention Delta<hr />Base model: <a target="_blank" rel="ugc" href="https://huggingface.co/latent-space-dreams/LS_Pingo_XL">latent-space-dreams/LS_Pingo_XL</a>Sampling: Uniform LogSNRSchedule: v-prediction &amp; ZTSNRDataset: ~10,000 curated SFW/NSFW anime imagesTrainer: <a target="_blank" rel="ugc" href="https://github.com/geltz/sdxl-trainer">sdxl-trainer</a> (Modified from <a target="_blank" rel="ugc" href="https://github.com/Hysocs/Aozora_SDXL_Training">Aozora</a>)<hr />License<a target="_blank" rel="ugc" href="https://freedevproject.org/faipl-1.0-sd/">FAIPL-1.0-SD</a>

Files with SHA256: 36f44e5aaab607d616719f9c1d59cef58ebe2a21e244e294503a0dc48b4cb1cb