Type | |
Stats | 507 782 |
Reviews | (89) |
Published | Sep 16, 2023 |
Base Model | |
Hash | AutoV2 E3325680B1 |
欢迎使用这个lora模型 这个模型 通过不同的底模可以实现2.5d风格或者类似游戏渲染模型的风格模型整体稍微偏向于2.5d 推荐使用 chilloutmix_NiPrunedFp32Fix ，Basil mix 这两个大模型 或者其他真人大模型 可以实现渲染CG风格 搭配dreamshaper_7这个大模型可以实现欧美网游的渲染风格
Welcome to use this lora model which can be implemented in a 2.5d style ora style similar to the game rendering model with different base modes. The overall model is slightly more favorable to 2.5d. chilloutmix_NiPrunedFp32Fix is recommended. Basil mix these two large models or other full-size models can be used to render CG style with dreamshaper_7 which is a large model to achieve the rendering style of European and American online games
------------------------------------------------------------------------------------------
使用建议（ Recommendations of use）:
推荐配真人模型使用兼容2.5d模型 使用类似真人的lora会有意外效果
It is recommended to use compatible 2.5d models using lora that resembles a real person will have unexpected effects
建议起手提示词（Suggest hand prompts）：best quality,masterpiece,realistic,HDR,UHD,8K,masterpiece,high quality,exquisite wallpaper,16k,animation,illustration,(white_skin),1girl,
------------------------------------------------------------------------------------------
推荐的参数（Recommended parameters）:
lora权重建议0.8 推荐开启AD修脸
采样方法（Sampler:）:
Euler a: 20~30步。
DPM++ 2S a Karras：25~30步。
DPM++ SDE Karras:20~30步。
DPM++ 2M SDE Karras：20~40步。
提示词引导系数(CFG Scale):7
人像人物分辨率（Character resolution）：512x768, 768x768，512x1024
高清修复（Hires.fix）：
放大算法:R-ESRGAN 4x+Anime6b ，4x-UltraSharp，4x_SmolFace_200k
高分迭代步数：10~15
重绘幅度：0.3~ 0.5
Clip Skip:2。
负面（Negatives): (worst quality, low quality:2),monochrome, zombie, (interlocked fingers:1.2),
------------------------------------------------------------------------------------------
Ai技术交流QQ群:377209248
图像源和训练代码来自互联网，模型仅用于科研兴趣交流。如有侵权，请联系删除，谢谢。
The image source and training code are from the Internet, and the model is only used for scientific interest exchange. If there is infringement, please contact to delete, thank you