这是第二个正式发布SDXLlora后续模型有什么问题可以向我反馈我会在后续版本中迭代

This is the second official release of the SDXLlora follow-up model, any questions can you give me feedback, I will iterate in the subsequent version

模型反馈技术交流QQ群:377209248

建议出图参数（Suggest drawing parameters）：

推荐SDXL底模（Recommended SDXL base mode）：StarlightXLAnimated_v2

LORA 建议权重0.6~0.8 LORA recommends a weight of 0.6 to 0.8

起手tag（Start tag）：best quality,masterpiece,realistic,full body,1girl,pink hair,k/da \(league of legends\),3D graphics,there is a star below the left eye,

采样方法（Sampler）:

DPM++ 3M SDE Exponential 45步及以上（推荐这个recommend）

DPM++ 2M SDE Karras 45步及以上

可以参考我下图设置Please refer to my Settings below

web/image/01101c247378369784f25b7d155704b21cd438081e89f2e2e7a915b788f342a5.png

切记低配置的电脑要开启1:1倍率高清修复Remember to turn on 1:1 HD repair for low-profile computers

提示词引导系数 (CFG Scale): 3~7

等倍高请修复人像人物分辨率（When times higher please fix portrait figure resolution）： 768x1280 ，1024x1024

高清修复（Hires.fix）: 4x-UltraSharp , 高分迭代步数 10~15 重绘幅度 0.3~ 0.5

Clip Skip: 2

负面提示词(Negative)：text,username,logo,(low quality, worst quality:1.4),(bad anatomy),(inaccurate limb:1.2),bad composition,inaccurate eyes,extra digit,fewer digits,(extra arms:1.2),watermark,multiple moles,mole on body,

本模型中不包含任何真人素材图像源和训练代码来自互联网，模型仅用于科研兴趣交流

This model does not include any real-life material, image sources, and training codes from the internet. The model is only used for scientific research interest exchange

KDA Girl-XL - v1.0

KDA Girl-XL - v1.0

Sampler

Steps

CFG Scale

Seed

Clip Skip

Size