Li Muwan's first edition generalized version

1. Notes on Image Production:

The original version I created has a good balance between fidelity and generalization. The generalized version may not be as faithful, but the skin and clothes will appear smoother. I recommend using the original version.

When refining, I added some things, those who understand, understand.

Because I am a bit lazy, the general models do not have information. However, I will post some images made with each model below. If needed, you can directly refer to the information below the image to draw.

1: Training material is 512*512, I recommend that everyone uses this model for image production with a size not exceeding 1024*1024. To achieve similar effects, you must use the base model I used for training! And make sure to use negative prompt words!

2: Please refer to the weight of this Lora model from the cover image.

3: ※ For bust portraits, you can skip high-resolution restoration for good results. If you want to add details by turning on high-resolution restoration, use the R-ESRGAN4x+Anime6B model with a redrawing magnitude of around 0.3-0.4.

※ If you need full-body portraits, be sure to turn on high-resolution restoration with a redrawing magnitude of around 0.5.

※ For full-body portraits, it is recommended to activate the ADetailer for facial features!

If you like this model, feel free to start running it! Everyone's interaction, collection, likes, and returning of images are the greatest motivation and support for me. Thank you again to all supporters!

2. Disclaimer (2024 Edition):

1. Without the permission of the creator (me), do not sell this model on any website.

2. Without explicit written consent, do not profit from using this model on other platforms.