V3

Detailed Drawing Parameters:

Stacking complex quality words is generally not needed when using my model

Simply input the content you want, the more complex the description, the less likely it is what you want

Typical Parameter Examples:

Positive Keywords: girl

Negative Keywords: (worst quality:1.5),(low quality:1.2),lowres,bad anatomy,

*Important!Choose Output Image Resolution: 512x768, 768x512 is the best, if you need a 16:9 aspect ratio or even larger, ensure the short side is 512, such as 512x896

VAE: None       Sampling Method: Any mainstream method works   Sampling Steps: 20-30   CLIP: 2

*Important!Please Enable High-Resolution Restoration Hires.fix

Zoom Factor: 2x         Zoom Algorithm: 4x-UltraSharp or R-ESRGAN 4x+ Anime6B   High-Resolution Iteration Steps: 15-20

*Redrawing Intensity:0.2-0.4, if greater than 0.5, the image changes significantly

Guide Word Coefficient CFG: 7

Select models for Controlnet Tile, lineart, Depth, use these models together, adjust control weights as needed

The above instructions, if you have no idea what I'm talking about, please search for relevant beginner tutorials on Bilibili, learn and then use this model

If you like my model, feel free to like, save, run, download, and provide feedback

If you have any feedback on the model, feel free to leave a comment in the comment section or upload your problematic image to the feedback area with a description, I will answer in the image comment section

Finally, I wish you a happy time playing with it, I hope it helps you