1. Notes on Image Output:

The original version maintains a high level of fidelity and generalization, while the generalized version may not preserve the details as well but will result in smoother skin and clothing. I recommend using the original version.

When refining, some extras are added, and well, those who understand, understand.

As the author is a bit lazy, there is usually no information on the general model, but some images produced with the model are provided below each model. If needed, you can directly refer to the information below the image to draw.

1: Training material is 512*512. I recommend not exceeding 1024*1024 for the output size using this model for equivalent effects. Be sure to use the base model I trained with for the output! And be sure to use negative indication terms!

2: This lora weight refers to the cover image.

3: ※For half-length images, high-resolution restoration is optional and can achieve good results without it. If adding details with high-resolution restoration, use the model R-ESRGAN4x+Anime6B with a redrawing range of about 0.3-0.4

※If full-length images are needed, high-resolution restoration is recommended with a redrawing range selection of about 0.5,

※For full-length images, it is recommended to open ADetailer for facial details!

Enjoy using this model and start running with joy! Your interactions, collections, likes, and feedback motivate and support me the most. Thank you once again to all supporters!

2. Disclaimer (2024 Edition):

1. Do not sell this model on any website without the explicit permission of the creator (me).

2. Do not profit from using this model on other platforms without clear written consent.

Wang Qui'er - V1.0 general,,2