<h2>[中文]</h2>低模<a target="_blank" rel="ugc" href="https://civarchive.com/models/795765/illustrious-xl">Illustrious-XL</a>支持任何常规分辨率（垂直或纵横只要是64倍数下）权重推荐&lt;0.85&gt;至&lt;1&gt;基本触发词 = hagoromo_gitsune \(nurarihyon no mago\),1girl,solo,pale skin,black mature_eyes,no pupils,small breasts服装 = school uniform,black serafuku,black skirt,black pantyhose细节 = fox tail,multiple tails负面提示词 = shoes,hagoromo,(floating_hagoromo:1.1),fat,full black eyes,bags under eyes,floating hair,messy hair,panty focus,abstract,hime cut,multiple tails （*添加greyscale可以让肤色看起来“正常”）*可能有帮助的提示词 = close up,shaded face,simple background<h3>使用建议：</h3><ol><li>强烈建议把shoes放进neg 否则很高几率会生成质量如下图1</li><li>提高分辨率如 1152x1536 效果可能会更理想</li><li>可以稳定配合大部分LoRA一起使用（适当调低权重）</li><li>建议配合Hires+Adetailer效果更好</li></ol>*图1 （解释：因为镜头太远加上分辨率太小，要么close up 要么提高分辨率）<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/52a49345-f1e1-4f7a-8854-f85d86b1462c/width=525/52a49345-f1e1-4f7a-8854-f85d86b1462c.jpeg" /><h2 id="eng">[Eng]</h2>Trained with <a target="_blank" rel="ugc" href="https://civarchive.com/models/795765/illustrious-xl">Illustrious-XL</a> supports any standard resolution (vertical or horizontal, as long as it's a multiple of 64). Recommended weight: &lt;0.85&gt; to &lt;1&gt;.Basic trigger words = hagoromo_gitsune \(nurarihyon no mago\),mature female,solo,pale skin,black mature_eyes,no pupils,small breastsOutfit = school uniform,black serafuku,black skirt,black pantyhoseDetails = fox tail,multiple tailsNeg prompts = shoes,hagoromo,(floating_hagoromo:1.1),fat,full black eyes,bags under eyes,floating hair,messy hair,panty focus,abstract,hime cut,multiple tails (Adding “greyscale” can make the skin tone appear more “normal”.)*some prompts that maybe can help to improve quality = close up,shaded face,simple background<h3 id="advance:">Advance:</h3><ol><li>t’s strongly recommended to place “shoes” in the negative prompt. Otherwise, there’s a high chance it will generate results similar to Image 1.</li><li>Using a higher resolution like 1152x1536 can achieve better results.</li><li>This model can be stably combined with most LoRA models (reduce weight accordingly).</li><li>It is recommended to use Hires+Adetailer for better results</li></ol>Image 1 (Explanation: Because the camera distance is too far and the resolution is too low, you either need to use a close-up or increase the resolution.)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/52a49345-f1e1-4f7a-8854-f85d86b1462c/width=525/52a49345-f1e1-4f7a-8854-f85d86b1462c.jpeg" />

Files with SHA256: c324551949a9cf19ae15feb419fb9d865543db8ddc8e38c0b8c0a33c5dacb3c3