Random seed. Leave blank to randomize the seed

Low quality input image.

Classifier-free guidance scale for prompts.

Default

7.5

Minimum

1

Maximum

20

Original churn hy-param of EDM.

Default

5

Original noise hy-param of EDM.

Default

1.003

Upsampling ratio of given inputs.

Default

1

Additive positive prompt for the inputs.

Default

"Cinematic, High Contrast, highly detailed, taken using a Canon EOS R camera, hyper detailed photo - realistic maximum detail, 32k, Color Grading, ultra HD, extreme meticulous detailing, skin pore detailing, hyper sharpness, perfect without deformations."

Minimum resolution of output images.

Default

1024

Negative prompt for the inputs.

Default

"painting, oil painting, illustration, drawing, art, sketch, oil painting, cartoon, CG Style, 3D render, unreal engine, blurring, dirty, messy, worst quality, low quality, frames, watermark, signature, jpeg artifacts, deformed, lowres, over-smooth"

Control Strength of Stage1 (negative means invalid).

Default

-1

Control Strength of Stage2.

Default

1

Number of steps for EDM Sampling Schedule.

Default

50

Minimum

1

Maximum

500

Use LLaVA model to get captions.

Default

true

Linearly (with sigma) increase CFG from 'spt_linear_CFG' to s_cfg.

Choose a model. SUPIR-v0Q is the default training settings with paper. SUPIR-v0F is high generalization and high image quality in most cases. Training with light degradation settings. Stage1 encoder of SUPIR-v0F remains more details when facing light degradations.

Default

"SUPIR-v0Q"

Color Fixing Type..

Default

"Wavelet"

Start point of linearly increasing CFG.

Default

1

Linearly (with sigma) increase s_stage2 from 'spt_linear_s_stage2' to s_stage2.

Start point of linearly increasing s_stage2.