QQ|Archiver|手机版|小黑屋|AiOMG ( 粤ICP备2022147020号-1 )

GMT+8, 2024-11-24 10:05 , Processed in 0.037308 second(s), 29 queries .

Powered by Discuz! X3.5 Licensed

© 2001-2024 Discuz! Team.

[LoRA] LingYi_Face

27
0
0
0
simonlive LV9
正文
发布时间:2024-08-02
模型类型
模型类型: LoRA
基础模型: SD 1.5
最新版本: v1.0
触发词:
下载链接: https://pan.baidu.com/s/1juxvciBKdXkx_VSIE9wCCg?pwd=5gv5
提取码: 5gv5
源链接: https://civitai.com/models/90090?modelVersionId=95935
模型简介:
脸部 Lora 模型,基于 majicMIX realistic V6 训练。
与 majicMIX realistic 或 majicMIX sombre 搭配使用,推荐参数:
模型权重: 0.4~0.6(过高的权重会降低质量)
提示词引导系数 CFG Scale: 5~7
采样步数 Sampling steps: 15~25
可以通过分层控制减少污染,例如:1,0,0,0,0,0,0,0,1,1,1,0,0,0,0,0,0,分层控制可以尝试更高的模型权重。
A human face Lora model, trained based on MajicMIX realistic V6.
Used in conjunction with majicMIX realistic or majicMIX symbol
Excessive model weights can reduce quality, and pollution can be reduced through hierarchical control, such as 1,0,0,0,0,0,0,0,1,1,1,0,0,0,0,0,0. Hierarchical control can try higher model weights.

~~~~~~~~~~~~~~~~~~~~~~~o(* ̄︶ ̄*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数:
Prompt: best quality,masterpiece,ultra high res,(photorealistic:1.4),1girl,<lora:LingYi_V1:0.55>,
Negative prompt: easynegative,ng_deepnegative_v1_75t,Jeans,
Steps: 20, CFG scale: 5, Sampler: Euler, Seed: 3970714652

本帖子中包含更多资源

您需要 登录 才可以下载或查看,没有帐号?立即注册 微信登录 手机动态码快速登录

x
回复

使用道具

 
您需要登录后才可以回帖 登录 | 立即注册 微信登录 手机动态码快速登录

本版积分规则

快速回复 返回顶部 返回列表