VRChat Manuka

Trained based on animagineXLV3_v30, generate 3D model images of Manuka for sale on the booth.

Recommended weight: 1;

The dataset is trained at a resolution of 1152*1152, comprising 68 images.

Trigger words: manuka, apron.

The labeling method involves removing the character's distinctive features.

If the image tag does not take effect, you can try to remove the trigger word, not writing the trigger word also has a certain guiding effect.

The simplified version has a slight quality degradation.

=========================================================

基于animagineXLV3_v30训练，生成Booth上售卖的3D模型Manuka图像。

推荐权重：推荐1；

数据集以1152*1152分辨率进行训练，有68张。

触发词：manuka，apron。

打标方式是删除人物形象特征。

如果形象tag没有生效，可以尝试去掉触发词，不写触发词有一定的引导效果

简化版本有轻微质量下降。