<a target="_blank" rel="ugc" href="https://civarchive.com/bounties">Introduction:</a>"Yatogami Tohka " is a character from the anime "Date a Live". This Model is trained on SDXL / Pony, This is a SFW Model, Any Kind of NSFW is strictly Not Allowed. This Model is Capable for Perfect Face and Casual wear.<hr />Recommendations:<ul><li>Use danbooru tagging for prompts</li><li>Sampler: "Euler a" at / 25 steps</li><li>Cfg scale: 8</li><li>Clip Skip: 2 &lt;---- VERY IMPORTANT</li><li>VAE : sdxl_vae</li><li>Add Before Prompt : <code>score_9,score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, score_anime, score_realstic</code></li><li>Negatives: <code>(EasyNegative:0.8), (worst quality, low quality:1.2), blurry, bokeh, dof, fog</code></li></ul>Recommendations (Hires Fix):<ul><li>Upscaler: <a target="_blank" rel="ugc" href="https://huggingface.co/FacehugmanIII/4x_foolhardy_Remacri/resolve/main/4x_foolhardy_Remacri.pth">Remacri</a></li><li>Hires steps: 10-15</li><li>Denoising strength: 0.3</li><li>Upscale by: any</li></ul><hr />If you want to get mostly the same results, you definitely will need negative embedding:<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/7808/easynegative">EasyNegative</a>, it's better to use it at 0.8 weight.</li></ul>Don't use a lot negative embeddings at the same time. It definitely will ruin any possible output on all models! If you want to use negative embeddings - use one for quality, and one for hands (if you want a placebo for hands).

Files with SHA256: 048f9f5826cfe0be53e099219f14da58403722f2d03eb9fd334c1981998b2749