2.0版本更新
使用了1000张高质量的图片训练效果非常棒！
2.0 version update
The training effect was great using 1000 high-quality images!
1.5版本：
更小的体积，更好的画面，更自由的动作和换装
此版本克服了1.0版本的动作过拟,并具有更高的自由度。需要注意的是，该版本可能在角色的眼睛、服装或其他一些人物特征上还不够还原，这将在下一个版本中得到改进。
Version 1.5:
Smaller size, better graphics, more freedom in movements and customization.
This version overcomes the exaggerated movements of version 1.0 and offers greater freedom. It is worth noting that this version may not fully capture the character's eyes, clothing, or other features, but these aspects will be optimized in the next version.
训练模型：SDXL1.0官模
测试模型：Kohaku-XL
建议权重：0.8 or adjust them freely according to the situation.
Training model: SDXL1.0 Official Model
Testing model: Kohaku-XL
Recommended weights: 0.8 or adjust them freely according to the situation.
Description
FAQ
Comments (2)
加载lora会使线条变粗？，同seed同参数下跑图，仅大模型线条会更细，是普遍现象吗
这个并不是预期，我个人猜测lora似乎把原素材的画风写进去了，目前我们还在研究，这有待在未来的版本中得到解决
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.