林襄 Lin Xiang

Don't know why the page becomes 404, so repost

-

Using the Chilloutmix-Ni 7G model, the training LORA model has 46 photos and 30 steps

-

Model: Chilloutmix-Ni,

Steps: 30,

Sampler: DPM++ 2M Karras,

CFG scale: 12,

Face restoration: CodeFormer,

Size: 512x768,

-

Denoising strength: 0.6,

Hires upscale: 2,

Hires steps: 30,

Hires upscaler: Latent (nearest-exact)

-

It is recommended to enable Highres.fix