<h3>2023/4/27 update</h3>Update dataset. Thanks <a target="_blank" rel="ugc" href="https://civarchive.com/user/hellofavgav">hellofavgav</a> for providing images.<h1>Introduction</h1>It's my first time to train a model about a single feature of different characters. This LoRA model is trained with ~130 images and can be used to generate different characters with black sclera. The trigger tag is 'black sclera, colored sclera'. The 784mb VAEs (NovelAI, Anythingv3, <a target="_blank" rel="ugc" href="https://huggingface.co/WarriorMama777/OrangeMixs">Orangemix</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/4468/counterfeit-v25">Counterfeit</a>) are recommended. 0.6 ~ 0.8 weights are good. The preview images are generated using 1.6 version (original) <a target="_blank" rel="ugc" href="https://civarchive.com/models/9139/yesmix">YesMix.</a><h1>Cautions (Must Read)</h1><ul><li>Some preview images are generated using 'Hires Fix' and upscaled using 'SD Upscale'. Thus, if you wanna generate the preview images, pls download the original preview images and analyze them using 'PNG INFO' in webui to get all generating information.</li><li>The maximum of generating resolution should not exceed 1024 theoretically, otherwise extra/deformed head/limbs will be waiting for you :). Try to use 'hires fix' to prevent this problem and help to generate higher-resolution images.</li><li>For 784mb VAEs (NovelAI, Anythingv3, Orangemix, Counterfeit) users, make sure add '--no-half-vae' in command line of webui to prevent generating black images. See <a target="_blank" rel="ugc" href="https://rentry.org/voldy">voldy's log</a> for the details of edit method. For high-performance GPU, make sure 'medvram' and 'lowvram' mode are disabled.</li></ul>

Files with SHA256: 7c0ad36445f9e08e9acc8e6c8443784b17b42b19fd97ec9596c8552349e14355