Model Description

This LoRA is fine-tuned on a dataset of 300+ images. It can recreate Yaoguang using the trigger word Yaogaung alone, while also providing detachable clothing tags.

In addition to LoRA based on NoobAI IllustriousXL, I have also trained versions based on IllustriousXL, Anima, AnimagineXL, PonyXL, and WAI-IllustriousXL.

Usage Tips

• Clothing: Special optimization has been applied to address the "hard-baked" clothing issue. Yaoguang's official outfit is now detachable. Her official design is primarily bound to the following tags: dress, bracelet, hair ornament.

• 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 (e.g., dress, hair ornament, etc.) to the negative prompt.

Recommended Settings

• Recommended weight: 0.7 - 1.0

• (Please adjust the prompts according to the official documentation of the checkpoint model you are using.)

• upper-body prompt :

masterpiece, best quality, very aesthetic, Yaoguang, 1girl, solo, medium breasts, hair ornament, bracelet, dress, bare shoulders, looking at viewer, smile, confident, sideboob, upper body, romantic, love, detailed light, epic, dramatic,

• SilvermoonMix-Illustrious prompt :

blurry background, (close-up:1.4), (foreshortening:1.4),from side, cowboy shot, (leaning forward, leaning to the side:1.2), white border, simple background, blue background, perspective, Yaoguang, 1girl, solo, adult, long hair, medium breasts, hair ornament, bracelet, feather, dress, bare shoulders, looking at viewer, smiling, cute, playful, dynamic, romantic, love, detailed light,

For full-body shots, you can try adding full body and high heels.

Others

While I am very satisfied with the accuracy of Yaoguang's features, I cannot guarantee that every detail is perfectly stable yet. I plan to further optimize the details after the official 3D model is released.

If you have any feedback, please feel free to leave a comment.

Thank you for your support!

模型简介

v1.1有最高还原度

300+ 张图片的小数据集上微调得到。此LoRA使用触发词Yaoguang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。

除了基于NoobAI的版本，我还做了基于IllustriousXL, Anima, AnimagineXL, PonyXL, WAI-IllustriousXL, and ZukiAnimeIll的版本。

使用建议

• 服装与配饰 ： 针对服装 hard-baked问题进行了专门优化。现在爻光的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：dress, bracelet, hair oenament

• 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（如 dress, hair ornament 等）加入到负面提示词中。

推荐设置

• 推荐权重： 0.7 - 1.0

• （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。）

• 上本身prompt:

masterpiece, best quality, very aesthetic, Yaoguang, 1girl, solo, medium breasts, hair ornament, bracelet, dress, bare shoulders, looking at viewer, smile, confident, sideboob, upper body, romantic, love, detailed light, epic, dramatic,

• SilvermoonMix-Illustrious 参考prompt :

blurry background, (close-up:1.4), (foreshortening:1.4),from side, cowboy shot, (leaning forward, leaning to the side:1.2), white border, simple background, blue background, perspective, Yaoguang, 1girl, solo, adult, long hair, medium breasts, hair ornament, bracelet, feather, dress, bare shoulders, looking at viewer, smiling, cute, playful, dynamic, romantic, love, detailed light,

全身可尝试添加full body, high heels 等。

小记

虽然我已经对爻光的特征还原程度很满意，但不能保证所有细节稳定，我会在官方3d模型发布后针对细节优化。

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：https://space.bilibili.com/12595237?spm_id_from=333.1007.0.0

小红书账号：https://www.xiaohongshu.com/user/profile/670412f2000000001d0209cd