Xiaoxuner new building model generalization version

1. Drawing Notes:

The original version has good fidelity and generalization, while the generalized version may not restore as well but will have smoother skin and clothing. I recommend using the original version.

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

Because the author is somewhat lazy, general models may lack information, but there will be some images made with the model under each model. You can directly refer to the information below the images to draw.

1: Training material is 512*512, I recommend not exceeding 1024*1024 for the output size using this model. To achieve the same effects, you must use the base model I trained with and use negative prompting words!

2: The Lora weight reference the cover image.

3: ※For half-body portraits, high-resolution repair is not necessary, and it will have good effects. If you want to enhance detail with high-resolution repair, use the model R-ESRGAN4x+Anime6B with a repaint amplitude of about 0.3-0.4.

※For full-body portraits, high-resolution repair is necessary at an amplitude of around 0.5,

※It is recommended to open ADetailer for portraits!

If you like this model, feel free to start running it! Your interaction, collection, likes, and feedback are my greatest motivation and support. Thank you once again to all supporters!

2. Disclaimer (Dec. 2024):

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

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