Cao Ying Generalization Test

1. Notes on Image Production:

The original version has good fidelity and generalization, while the generalized version may not be very faithful but the skin and clothing will look smoother. I recommend using the original version.

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

Because the author is lazy, there is usually no information about the models. However, some example images created by each model will be posted below. Those who need them can directly refer to the information below to draw.

1: Training material size is 512*512, it is recommended that the output image size for this model does not exceed 1024*1024. For the same effect, you must use the base model that I trained with! And be sure to use negative prompts!

2: Use the Lora weight reference cover image for this model.

3: Semi-portraits may not require high-resolution repair for good results. To add detail, use high-resolution repair with the model R-ESRGAN4x+Anime6B, with a retouching magnitude of around 0.3-0.4.

For full-body portraits, high-resolution repair is recommended, with a retouching magnitude of around 0.5.

Full-body portraits are recommended to enable ADetailer for the facial area!

If you like this model, feel free to start running it! Everyone's interaction, collection, likes, and feedback are the greatest motivation and support for me, I thank all the supporters once again!

2. Disclaimer (2024):

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

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