<h1 id="model-description">Model Description</h1>v1.3 has the highest fidelity.This LoRA is fine-tuned on a dataset of 300+ images. It can recreate Sparxie using the trigger word alone, while also providing detachable clothing tags for creative flexibility.In addition to LoRA based on IllustriousXL, I've also trained versions based on Anima, Flux Klein 9b &amp; 4b, NoobAI, AnimagineXL, PonyXL and WAI-IllustriousXL. Check in my profile.<h1 id="recommended-settings">Recommended Settings</h1><ul><li>Recommended weight: 0.8 - 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, Sparxie, 1girl, hat, smile, looking at viewer, choker, cowboy shot, leg up, standing, cute, playful, dark, late hours, simple background, highly detailed,</code></pre>For full-body shots, you can try adding <code>full body</code> and <code>boots</code>.<h1 id="usage-tips">Usage Tips</h1><ul><li>Clothing &amp; Accessories: Sparxie's official outfit is detachable. Her official design is primarily bound to the following tags: <code>hat, dress, choker, gloves, boots, earrings</code>.</li><li>Changing Outfits: In my testing with common checkpoint models, removing the official clothing tags allowed for adding any preferred clothing elements. If residual elements persist when using your specific checkpoint for inference, consider adding the aforementioned tags to the negative prompt.</li></ul><h1 id="others">Others</h1><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Models are automatically deployed via AI Agent. Please report any parameter errors or version issues in the comments.While I am satisfied with the fidelity of features like her facial mark, mini hat, earrings...please feel free to leave your generations or feedback.Thank you for your support!<hr /><h1>模型简介</h1>v1.3 具有最高的还原度。这个LoRA模型是角色火花（Sparxie）的角色增强模型，在300+ 张图片的数据集微调得到。此LoRA使用触发词即可还原人物，也提供了可拆卸的服装提示词，方便各类二创发挥。除了IllustriousXL, 我还微调了 Anima, Flux Klein 9b &amp; 4b, NoobAI, AnimagineXL, PonyXL 和 WAI-IllustriousXL的版本，见我主页。<h1>推荐设置</h1><ul><li>推荐权重： 0.8 - 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, Sparxie, 1girl, hat, smile, looking at viewer, choker, cowboy shot, leg up, standing, cute, playful, dark, late hours, simple background, highly detailed,</code></pre>全身可尝试添加<code>full body</code>, <code>boots</code> 等。<h1>使用建议</h1><ul><li>服装与配饰 ： 针对服装 hard-baked问题进行了专门优化。现在官方外设是可拆卸的，官方服设主要绑定在绑定了以下标签：<code>hat, dress, choker, gloves, boots, earrings</code>。</li><li>更换服装： 实测下来我在我常用的checkpoint模型上在去掉上述官方服装提示词后，可以自由添加任意喜欢的服装元素。如果你在使用其他checkpoint模型推理时仍发现有官方服装残留，建议将上述标签加入到负面提示词中。</li></ul><h1>小记</h1>模型由AI Agent全自动发布，若参数、模型版本有问题，欢迎评论反馈。虽然我已经对火花的特征还原程度很满意，包括眼部印记、小礼帽、耳饰等等，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账号：<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>小红书账号：<a target="_blank" rel="ugc" href="https://www.xiaohongshu.com/user/profile/68483ecb000000001b019555">https://www.xiaohongshu.com/user/profile/68483ecb000000001b019555</a>

Sparxie | 火花 (Detail Enhanced 细节增强) Honkai: Star Rail

Files with SHA256: 56fe8650c644d813a135b8e0e5b46f08c330e52b7d9b563a347d2b2b842ea537