The style of this version is different from the previous version. This version is not an iterative version but a branch version. You can choose according to your preference. This version adds elements of ancient Chinese style, a Chinese face shape, and elements of Hanfu. Version recommendations Recommended prompt words for this version: Note that this version has trigger words: Face trigger words: chinese girls Hanfu trigger word: hanfu Positive: Best quality, Chinese girls, Negative: (worst quality, low quality, blurry, bad eye, ), (wrong hand, bad anatomy, wrong anatomy, ), (cgi, illustration, cartoon, poorly drawn, watermark), head out of frame, Sampling method (sampler): dpm++ 3m sde exponential 45 steps and above dpm++ 3m sde karras 45 steps and above HD repair (hires.fix): 4x-ultrasharp, high score iteration steps 10~15, redraw range 0.3~ 0.5 clip skip: 2 ADetailer can be turned on for local image generation (do not enable it for Libu online image generation) Version 2.0 content This model is extracted and fused with lora from other sdxl base models, in order to allow people who are new to sdxl to adapt to sdxl faster and allow them to experience the fun of sdxl. This model is easier than the original sdxl Easier to get started. Wide compatibility. Very versatile model with built-in VAE. No need to add additional VAE models.