Dingdang fusion Dingdang Fusion V6
v1.0
Welcome to Use the Model (This Model is a Self-Fusion Model)
This model can generate a pseudo-3D effect similar to Chinese comics
Recommended parameters:
Sampling method (Sampler):
Euler a: 20~30 steps.
DPM++ 2S a Karras:25~30 steps.
DPM++ SDE Karras:20~30 steps.
DPM++ 2M SDE Karras:20~40 steps.
DDIM:30 steps and above.
Prompt word guide coefficient (CFG Scale):7
Portrait resolution (Character resolution):512x768, 768x768, 512x1024
High-definition restoration (Hires.fix):
Enlargement algorithm:R-ESRGAN 4x+Anime6b, 4x-UltraSharp, 4x_SmolFace_200k
High score iteration steps:10~15
Redraw extent:0.3~0.5
Clip Skip:2.
Negatives: (worst quality, low quality: 2), monochrome, zombie, (interlocked fingers: 1.2),
------------------------------------------------------------------------------------------
The image source and training code are from the Internet, the model is only used for academic research and communication of interest. If there is any infringement, please contact for deletion, thank you.