<h1 id="model-description">Model Description</h1>v1.0 has the highest fidelity.This LoRA is for Fluorite (萤石 / フローライト) from Arknights: Endfield, fine-tuned on a dataset of 500+ images. It can recreate Fluorite using only the trigger word <code>Fluorite</code> .In addition to LoRA based on IllustriousXL, I have also trained versions based on NoobAI, AnimagineXL, PonyXL, WAI-IllustriousXL. Coming soon.<h1 id="param-settings">Param Settings</h1><ul><li>Recommended weight: 0.7 - 1.0</li><li>example prompt :(Please adjust the prompts according to model card of the checkpoint you are using.)</li></ul><pre><code>masterpiece, best quality, amazing quality, Fluorite, 1girl, solo, cowboy shot, simple background, smile, closed mouth,</code></pre>For full-body shots, you can try adding <code>full body</code> or <code>boots</code>.<h1 id="tips">Tips</h1><ul><li>Official design is bound to the following tags: <code>jacket, gloves, pants, boots, tail</code>.</li><li>If you do not want the hood, add <code>hood, hood up</code> to the negative prompt. The default is a single ponytail.</li><li>If you do not want the tail, add <code>tail</code> to the negative prompt.</li><li>If you find the official outfit elements are not decoupling when using your own checkpoint, lower model weight or add <code>jacket, gloves, pants, boots</code> to the negative prompt.</li></ul><h1 id="others">Others</h1>While I am satisfied with the fidelity of features like her hairclip, bodysuit, and subtle expressions...please feel free to leave your generations or feedback.Thank you for your support!<hr /><h1>模型简介</h1>v1.0 有最高还原度。这个LoRA模型是萤石/Fluorite/フローライト（明日方舟：终末地）的角色增强模型，在 500+ 张的数据集微调得到。此LoRA仅使用触发词 <span style="background-color:rgba(135, 131, 120, 0.15);color:rgb(235, 87, 87);font-family:SFMono-Regular, Menlo, Consolas, &quot;PT Mono&quot;, &quot;Liberation Mono&quot;, Courier, monospace;font-size:85%;line-height:normal">Fluorite即可还原人物。除了基于Illustrious XL的版本，我还训练了基于 NoobAI，AnimagineXL，PonyXL，WAI-IllustriousXL 的版本。Coming soon.<h1>参数设置</h1><ul><li>推荐权重： 0.7 - 1.0</li><li>（请根据你实际使用的Checkpoint模型的官方文档，适当调整提示词策略。）</li><li>示例prompt:</li></ul><pre><code>masterpiece, best quality, amazing quality, Fluorite, 1girl, solo, cowboy shot, simple background, smile, closed mouth,</code></pre>全身可尝试添加<code>full body</code>, <code>boots</code> 等。<h1 id="tips">Tips</h1><ul><li>官方服设绑定在以下标签：<code>jacket, gloves, pants, boots, tail</code>。</li><li>如果你不想要兜帽，在负面提示词加 <code>hood, hood up</code>。默认单马尾。</li><li>如果你不想要尾巴，在负面提示词加 <code>tail</code>。</li><li>如果你在使用底模进行换装时，发现服装元素不能完全解耦，请在负面提示词加入 <code>jacket, gloves, pants, boots</code>。</li></ul><h1>小记</h1>虽然我已经对她的特征还原程度很满意，包括发夹、紧身衣、微妙の小表情等等，但如果有任何反馈意见欢迎评论，以便我做出调整，谢谢。我会定期在社交媒体发布图片，可在对应平台取原图。Bilibili：<a target="_blank" rel="ugc" href="https://space.bilibili.com/12595237?spm_id_from=333.1007.0.0">https://space.bilibili.com/12595237?spm_id_from=333.1007.0.0</a>

Files with SHA256: 20b1c71fc814bbcbba3020a680ec61cf3ffe8b9bf5a0c1801d21b9aa5ee5fd2b