This merge maintains Illustrious 0.1's output while stabilizing it and improving contrast. I use it to test LoRAs I train on the <a target="_blank" rel="ugc" href="https://civarchive.com/models/795765/illustrious-xl">base model</a>.models used <a target="_blank" rel="ugc" href="https://civarchive.com/models/811067/illustrious-xl-smoothft">illustrious-xl-smoothft_spo</a> <a target="_blank" rel="ugc" href="https://civarchive.com/models/246299/comradeship-xl">comradeship-xl_v14t1s</a> tool used <a target="_blank" rel="ugc" href="https://github.com/ljleb/sd-mecha">sd_mecha</a> <a target="_blank" rel="ugc" href="https://huggingface.co/reoe/illustrious0.1-smoothftspo-v14t1s-rotate/blob/main/recipe.py">(recipe)</a>

illustrious0.1-smoothftspo-v14t1s-rotate

<ol><li>weighted sum <a target="_blank" rel="ugc" href="https://civitai.com/models/246299/comradeship-xl">1</a> and <a target="_blank" rel="ugc" href="https://civitai.com/models/303060/jruthe-journey-remains-unfinished25">2</a> = result</li><li>weighted sum result with <a target="_blank" rel="ugc" href="https://civitai.com/models/303060/jruthe-journey-remains-unfinished25">this model</a></li></ol>

Files with SHA256: 1756e63de1f52fec1d0f595a85e47bdf1766b690a803a5289a48454dd2677f8f