lucataco/pasd-magnify 🔢🖼️📝 → 🖼️

▶️ 42.3K runs 📅 Jan 2024 ⚙️ Cog 0.8.6 🔗 GitHub 📄 Paper

About

(Academic and Non-commercial use only) Pixel-Aware Stable Diffusion for Realistic Image Super-resolution and Personalized Stylization

Example Output

Prompt:

"Frog, clean, high-resolution, 8k, best quality, masterpiece"

Output

Performance Metrics

13.32s Prediction Time

143.46s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/KBuUhAMqvkXfjTqqOyg1TqIXSzgUhCbjIzaFoNb7fUdkK685/frog.png",
  "prompt": "Frog, clean, high-resolution, 8k, best quality, masterpiece",
  "n_prompt": "dotted, noise, blur, lowres, oversmooth, longbody, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality",
  "denoise_steps": 20,
  "guidance_scale": 7.5,
  "upsample_scale": 2,
  "conditioning_scale": 1.1
}

Input Parameters

seed Type: integer: Random seed. Leave blank to randomize the seed
image (required) Type: string: Input image
prompt Type: stringDefault: Frog, clean, high-resolution, 8k, best quality, masterpiece: Prompt
n_prompt Type: stringDefault: dotted, noise, blur, lowres, oversmooth, longbody, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality: Negative Prompt
denoise_steps Type: integerDefault: 20Range: 10 - 50: Denoise Steps
guidance_scale Type: numberDefault: 7.5Range: 0.5 - 10: Guidance Scale
upsample_scale Type: integerDefault: 2Range: 1 - 4: Upsample Scale
conditioning_scale Type: numberDefault: 1.1Range: 0.5 - 1.5: Conditioning Scale

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Using seed: 2896644955
Seed set to 2896644955
  0%|          | 0/20 [00:00<?, ?it/s]
  5%|▌         | 1/20 [00:00<00:12,  1.54it/s]
 10%|█         | 2/20 [00:01<00:10,  1.76it/s]
 15%|█▌        | 3/20 [00:01<00:09,  1.82it/s]
 20%|██        | 4/20 [00:02<00:08,  1.86it/s]
 25%|██▌       | 5/20 [00:02<00:07,  1.89it/s]
 30%|███       | 6/20 [00:03<00:07,  1.90it/s]
 35%|███▌      | 7/20 [00:03<00:06,  1.91it/s]
 40%|████      | 8/20 [00:04<00:06,  1.91it/s]
 45%|████▌     | 9/20 [00:04<00:05,  1.92it/s]
 50%|█████     | 10/20 [00:05<00:05,  1.92it/s]
 55%|█████▌    | 11/20 [00:05<00:04,  1.92it/s]
 60%|██████    | 12/20 [00:06<00:04,  1.92it/s]
 65%|██████▌   | 13/20 [00:06<00:03,  1.92it/s]
 70%|███████   | 14/20 [00:07<00:03,  1.92it/s]
 75%|███████▌  | 15/20 [00:07<00:02,  1.93it/s]
 80%|████████  | 16/20 [00:08<00:02,  1.93it/s]
 85%|████████▌ | 17/20 [00:08<00:01,  1.93it/s]
 90%|█████████ | 18/20 [00:09<00:01,  1.93it/s]
 95%|█████████▌| 19/20 [00:09<00:00,  1.93it/s]
100%|██████████| 20/20 [00:10<00:00,  1.93it/s]
100%|██████████| 20/20 [00:10<00:00,  1.90it/s]
[Tiled VAE]: the input size is tiny and unnecessary to tile.
[Tiled VAE]: Done in 11.102s, max VRAM alloc 5803.197 MB

Version Details

Version ID: d59e83ee13c42b137aee558c483e3acc0a8ecdacb1444a7be48152f008dcc195
Version Created: January 8, 2024

Run on Replicate →