This LoRA model was trained based on XXMix_9realisticSDXL V1.0. It would be better to use XXMix_9realisticSDXL V1.0 as your base model when using this model.
This version of liuyifei-XL LoRA serves as a test iteration. If you have any suggestions or feedback, please don't hesitate to share your comments.
It is always a challenging endeavor to imitate genuine individuals, as true individuals themselves undergo changes as time passes. We would continue to improve the quality and robustness of this model.
It was found that using hires.fix would cause blurry artifacts, I am trying to figure out this problem.