此模型训练的是上半身的，因此推荐只生成上半身的图像，全身图片效果很大可能不会太好？建议以这些TAG为基础进行扩展：masterpiece, best quality,beautifly,1girl,cinematic angle, upper body,（预览图的TAG）
权重的话建议高一点0.6以上。
负面TAG建议有：long neck，因为暂时还没有解决原训练集有点长脖子的问题
预览图均是由以下参数生成（SD生成图的上传不了）：
负面TAG用一般的就行了没有什么要求
Steps: 40, Sampler: DPM++ SDE Karras, CFG scale: 7
除了sd2.1和sd1.5外的任何一种3D的模型都可以
=========================================================
This model trains the upper body, so it is recommended to generate only the upper body image, while the whole body image has a great effect, which may not be good??
It is suggested to extend the model based on these tags: Masterpiece, best quality, the beautifly, 1 girl, cinematic Angle, upper body,(TAG of preview image)
higher weights that if more than 0.6
(Negative prompt: long neck),Because we haven't solved the long neck problem in the original training set yet
Preview images are generated by the following parameters (SD generated images cannot be uploaded)：
A general Negative prompt will do
Steps: 40, Sampler: DPM++ SDE Karras, CFG scale: 7
Any 3D model except sd2.1 and sd1.5 will do
Description
此模型为loha，放在lora文件夹下使用。相比与前版本增强了些泛化性生成全身像还是有点效果，由于训练集的局限型，不过也就那样了...建议较高的权重。
This model is loha and is used in the folders of Lora . Compared to the previous version, the enhancement of generalization in generating full body images still has some effect, due to the limitations of the training set, but that's all Suggest a higher weight.
FAQ
Comments (2)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.