本质上是我的第一个sdxl模型的1.5版本使用anythingv3为底模https://huggingface.co/swl-models/anything-v3.0/resolve/main/Anything-V3.0.ckpt 推荐权重为1-0.7
如果想要复现实例图片风格的话使用a+b+c的tag
A：red background,yellow background,purple background,pink background,orange background,green background,blue background,aqua background
B:two-tone background,simple background
C:flat color(可加可不加）
想要侧脸使用：from side
example：
1girl, solo, blue background, two-tone background, simple background, flat color, raifu, dark red hair, straight hair, long hair, blue eyes, cat ears, white sweater ,<lora:CleanLineArt:0.8>
Essentially, this is version 1.5 of my first SDXL model, using AnythingV3 as the base model. https://huggingface.co/swl-models/anything-v3.0/resolve/main/Anything-V3.0.ckpt Recommended weights range from 1 to 0.7.
To replicate the example image style, use tags from categories a + b + c:
A: red background,yellow background,purple background,pink background,orange background,green background,blue background,aqua background
B: two-tone background, simple background
C: flat color (optional)
For a profile view, use: from side
example:
1girl, solo, blue background, two-tone background, simple background, flat color, raifu, dark red hair, straight hair, long hair, blue eyes, cat ears, white sweater ,<lora:CleanLineArt:0.8>