Model Introduction:

This is a mature women style portrait of Lora, aiming to create dreamlike graphic to meet special taste you needs~

Dataset has been remaked in v3.0 to improve image quality

Recommended settings:

1. Set Clip skip to 1 or 2;

2. The commonly used CFG range is 5-7;

3. Step number: 20-30;

4. Commonly used resolutions are 512x512 or 512X768;

5. Trigger word milf_ saga, use milf_saga_face in v3.0 when face not displayed, optional trigger like bun, hair bun;

6. It is recommended to use negative prompts such as (best quality: 2), (low quality: 2), and (normal quality: 2)

Best quality and other prompts to improve image quality;

7. It is recommended to use easynegative、ng_deepnegative_v1_75t, etc;

The images generated by different seeds have significant differences. Wishing you gacha the most beautiful women in your heart~

模型简介：

这是一个成熟阿姨风人像lora~

使用训练图片数量128，旨在打造梦幻级图像满足您的特殊xp~

v3.0已对数据集进行全新制作，提升了图像质量

推荐设置：

1.Clip skip设为1或2；

2.常用的CFG范围是5~7

3.step步数20~30；

4.常用分辨率512x512、512X768；

5.触发词milf_saga，在版本3.0中不出现头部时可使用milf_saga_face，可选触发词例如bun, hair bun；

6.建议使用(worst quality:2), (low quality:2), (normal quality:2)等negative prompt、

best quality等prompt提升画质；

7.建议使用easynegative、ng_deepnegative_v1_75t等；

不同seed生成出的图像差别比较大，祝您抽中您心中最美的阿姨~