EMBRACE THE DARKNESS (and Ganyu, that's mandatory).Make sure to use UI that supports v-prediction, otherwise you will get a noisy blob!<h1 id="laxharnoobai-xl-vpred-1.0-based."><a target="_blank" rel="ugc" href="https://huggingface.co/Laxhar/noobai-XL-Vpred-1.0/blob/main/NoobAI-XL-Vpred-v1.0.safetensors">Laxhar/noobai-XL-Vpred-1.0</a> based.</h1><h3 id="main-parameters:">Main parameters:</h3><ul><li>Steps: 28</li><li>CFG: 3-5</li><li>Sampler-scheduler: euler/euler a-normal, DDPM-exponential (better noses)</li><li>CFGZeroStar and sometimes Self-Attention Guidance: default for difficult backgrounds</li></ul>Try to BREAK prompts into parts by 75 tokens max, this goes for most models..<h3 id="positive-prompt:">Positive prompt:</h3><pre><code>number of characters (1girl/1boy), character name, artist prompt, scene/environment/camera angle, action, expression, items, quality prompts (masterpiece, best quality)</code></pre><ul><li>some heavily stylized characters might require prompt: realistic, nose</li></ul><h3 id="negative-prompt:">Negative prompt:</h3><pre><code>lowres, worst quality</code></pre><ul><li>even longer negative should work, but I prefer simplicity.. tags worth noting include sketch, abstract, url</li></ul>Enjoy..

Soft Surrender vPred

Finally! Hehe~ Reworked dataset, chose only blocks responsible for what i like, global weights 0.9 to keep it from overwriting everything.. Mwah.

Files with SHA256: ec20f529a43178ab3a0dea8c1f69c8822848ce013fabd963c603d4cdeb4570f5