Original Version of Xiaobai

1. Attention to Drawing:

When refining, some things are added, but well, those who understand it, understand it.

As the author is quite lazy, generally, the models do not have information, but some images made using the model will be posted below each model. If needed, you can directly refer to the information below the images to draw.

1. Trained materials should be 512*512, we recommend not exceeding the image size of 1024*1024 with this model. For the same effect, you must use the base model I used in training! Also, be sure to use negative prompt words!

2. The lora weight should reference the cover picture.

3. ※ For half-body portraits, you can skip the high-resolution restoration and still achieve good results. If adding details with high-resolution restoration, use the model R-ESRGAN4x+Anime6B with a redraw magnitude of around 0.3-0.4.

※ If a full-body portrait is needed, high-resolution restoration must be used, with a redraw magnitude of around 0.5.

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

If you like this model, feel free to start running it! Your interactive actions such as saving, liking, and sharing back the images are my greatest motivation and support. Thanks again to all the supporters!

2. Disclaimer (2024):

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

2. Without explicit written consent, do not use this model to profit on other platforms.