KenCanMix
This merged model is between 2.8D to realistic.
VAE is required: https://huggingface.co/stabilityai/sd-vae-ft-mse-original
Hires upscale + Dynamic Thresholding is Highly Recommended.
Recommended Parameters*:
Sampler: DPM++ SDE Karras, DPM++ 2M Karras
Steps: 25-50
Hires upscaler: Latent (bicubic antialiased), ESRGAN_4x, R-ESRGAN 4x+
Hires upscale: 1.5+
Hires steps: 15+
Denoising strength: 0.4-0.6
CFG scale: 7-10
or
CFG scale: 20-30 + Dynamic Thresholding
Mimic scale: 7-10
Threshold percentile: 99+
Mimic mode: Half Cosine Up
Mimic scale minimum: 3-7
CFG mode: Half Cosine Up
CFG scale minimum: 4-7
Clip skip: 2
*This is just how I use so far, feel free to play around and share with us any better one.
===v.1.0===
If you want Asian faces, LoRA is MUST, and the weight should be higher, try using [from:to:when] prompts. Or applying embeddings such as Ulzzang-6500
Credit to the authors of the following:
Description
KenCanMix v2.0 beta
Not sure how to define this version but sometimes it could be interesting. It is not a completed version but sharing with the community.
This generates rather "random" results or unstable between 28D ~ realistic.
- Sharpnesss, light and shadow can be strong enough based on your prompts such as "detailed shadow, sharp focus, saturated, high contrast, strong light and shadow" etc.
- more realistic on skin, wrinkle, breast shape, skin spots, etc. If you do not like it, negative prompts should work.
- body curves / proportions could be plumper.
- more matured faces (if CFG>15).