lucataco/rivers-stable-diffusion-upscaler 🔢🖼️📝❓ → 🖼️

▶️ 455 runs 📅 Jul 2023 ⚙️ Cog 0.7.2 🔗 GitHub 📄 Paper
image-upscaling

About

RiversHaveWings Stable Diffusion Upscaler

Example Output

Prompt:

"the temple of fire by Ross Tran and Gerardo Dottori, oil on canvas"

Output

Example output

Performance Metrics

23.70s Prediction Time
23.67s Total Time
All Input Parameters 
{
  "eta": 1,
  "image": "https://replicate.delivery/pbxt/JC0Wu2il27rHPIWbtVxx0xaZu9IAD11PZMarH9Drc6pFC4rX/demo.jpeg",
  "steps": 50,
  "prompt": "the temple of fire by Ross Tran and Gerardo Dottori, oil on canvas",
  "sampler": "k_dpm_adaptive",
  "tol_scale": 0.25,
  "guidance_scale": 1,
  "noise_aug_type": "gaussian",
  "noise_aug_level": 0
}
Input Parameters
eta Type: numberDefault: 1Range: 0 - 10
Amount of noise to add per step (0-deterministic). Used in all samples except k_euler
seed Type: integerDefault: 0
Seed (0 = random, maximum: 2147483647)
image (required) Type: string
Input image
steps Type: integerDefault: 50Range: 0 - 100
# inference steps
prompt Type: stringDefault: the temple of fire by Ross Tran and Gerardo Dottori, oil on canvas
sampler Default: k_dpm_adaptive
Sampler to use
tol_scale Type: numberDefault: 0.25Range: 0 - 10
k_dpm_adaptive uses an an adaptive solver with error tolerance tol_scale, all other use a fixed # of steps
guidance_scale Type: numberDefault: 1Range: 0 - 10
Guidance scale (0 - 10)
noise_aug_type Default: gaussian
Sampler to use
noise_aug_level Type: numberDefault: 0Range: 0 - 1
Amount of noise to augment
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
No seed was provided, using the current time.
Generating with seed=1689660281
Global seed set to 1689660281
Version Details
Version ID
d70507adad3ca5f6723bbd289706b6ef9460bc0cd24d3f91db729a7695800cc4
Version Created
July 18, 2023
Run on Replicate →