<p><span style="color:rgb(34, 139, 230)"><em>A few nice images and video got hidden automatically — just turn on all the Browsing Levels to see them.</em></span></p><p></p><h1 id="v2">V2 </h1><h2 id="recommended-weight:-0.9-10.9-couldn't-be-better"><strong>Recommended weight: 0.9-1（</strong>0.9 couldn’t be better）</h2><p></p><h2 id="recommended-settings"><strong>Recommended settings</strong></h2><p><strong>Sampler:</strong> Euler</p><p><strong>Schedule type: </strong>Simple</p><p><strong>Steps: 30</strong></p><p><strong>CFG: </strong>5</p><p><strong>Hires. Fix:</strong> R-ESRGAN 4x + Anime6B</p><p><strong>Magnification:</strong> 2</p><p></p><h3 id="prompts"><strong>Prompts</strong></h3><h3 id="prompt-prefix:"><strong>Prompt Prefix:</strong></h3><pre><code>masterpiece, best quality, newest, absurdres, highres, safe, s2d3o, 1girl, solo, looking at viewer, bare shoulders, hair between eyes, smile, blush</code></pre><h3 id="negative-prompt:"><strong>Negative Prompt:</strong></h3><pre><code>nsfw, worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthro</code></pre><p></p><h1 id="v1">V1</h1><h2 id="recommended-weight:-0.8-0.9from-testing-0.8-seems-to-be-the-optimal-choice-it's-more-stable"><strong>Recommended weight: 0.8-0.9（</strong>From testing, 0.8 seems to be the optimal choice — it’s more stable）</h2><p></p><h2 id="recommended-settings"><strong>Recommended settings</strong></h2><p><strong>Sampler:</strong> Euler A</p><p><strong>Schedule type: </strong>Simple</p><p><strong>Steps: 30</strong></p><p><strong>CFG: </strong>6~7（It feels like the colors in 6 are richer）</p><p><strong>Hires. Fix:</strong> R-ESRGAN 4x + Anime6B</p><p><strong>Magnification:</strong> 2</p><p></p><h3 id="prompts"><strong>Prompts</strong></h3><h3 id="prompt-prefix:"><strong>Prompt Prefix:</strong></h3><pre><code>s2d3o, 1girl, solo, looking at viewer, bare shoulders, hair between eyes, smile,</code></pre><h3 id="negative-prompt:"><strong>Negative Prompt:</strong></h3><pre><code>nsfw,bad quality,worst quality,worst detail,sketch,censor,</code></pre><p></p>

Sara Da Odin

<p>This version train on NoobAI-XL (NAI-XL) V-Pred-1.0. I also optimized the dataset and improved the tagging strategy so the tags align better with the image features for the AI to learn,at the same time, i’m also exploring new training strategies</p>

Files with SHA256: 61c37d002948d5b9971c1933b0f65d331b04234dae661ef7795a59be464f4939