cjwbw/stable-diffusion 🖼️🔢❓📝 → 🖼️

▶️ 65.4K runs 📅 Aug 2022 ⚙️ Cog 0.4.1 🔗 GitHub ⚖️ License
image-inpainting image-to-image text-to-image

About

stable-diffusion with negative prompts, more scheduler

Example Output

Prompt:

"Ragdoll cat king wearing a golden crown, intricate, elegant, highly detailed, centered, digital painting, artstation, concept art, smooth, sharp focus, illustration, artgerm, Tomasz Alen Kopera, Peter Mohrbacher, donato giancola, Joseph Christian Leyendecker, WLOP, Boris Vallejo"

Output

Example output

Performance Metrics

2.60s Prediction Time
2.63s Total Time
All Input Parameters 
{
  "width": "512",
  "height": "512",
  "prompt": "Ragdoll cat king wearing a golden crown, intricate, elegant, highly detailed, centered, digital painting, artstation, concept art, smooth, sharp focus, illustration, artgerm, Tomasz Alen Kopera, Peter Mohrbacher, donato giancola, Joseph Christian Leyendecker, WLOP, Boris Vallejo",
  "scheduler": "K_EULER_ANCESTRAL",
  "num_outputs": 1,
  "guidance_scale": 7.5,
  "negative_prompt": "yellow color, nose",
  "prompt_strength": 0.8,
  "num_inference_steps": "30"
}
Input Parameters
mask Type: string
Black and white image to use as mask for inpainting over init_image. Black pixels are inpainted and white pixels are preserved. Experimental feature, tends to work better with prompt strength of 0.5-0.7
seed Type: integer
Random seed. Leave blank to randomize the seed
width Default: 512
Width of output image. Maximum size is 1024x768 or 768x1024 because of memory limits
height Default: 512
Height of output image. Maximum size is 1024x768 or 768x1024 because of memory limits
prompt Type: stringDefault:
Input prompt
scheduler Default: pndm
Choose a scheduer
init_image Type: string
Inital image to generate variations of. Will be resized to the specified width and height
num_outputs Default: 1
Number of images to output. NSFW filter in enabled, so you may get fewer outputs than requested if flagged
guidance_scale Type: numberDefault: 7.5Range: 1 - 20
Scale for classifier-free guidance
prompt_strength Type: numberDefault: 0.8
Prompt strength when using init image. 1.0 corresponds to full destruction of information in init image
num_inference_steps Type: integerDefault: 50Range: 1 - 500
Number of denoising steps
Output Schema

Output

Type: array • Items Type: string • Items Format: uri

Example Execution Logs 
Using seed: 7556
  0%|          | 0/30 [00:00<?, ?it/s]
  7%|▋         | 2/30 [00:00<00:02, 13.46it/s]
 13%|█▎        | 4/30 [00:00<00:01, 14.22it/s]
 20%|██        | 6/30 [00:00<00:01, 14.00it/s]
 27%|██▋       | 8/30 [00:00<00:01, 14.31it/s]
 33%|███▎      | 10/30 [00:00<00:01, 14.43it/s]
 40%|████      | 12/30 [00:00<00:01, 14.61it/s]
 47%|████▋     | 14/30 [00:00<00:01, 14.36it/s]
 53%|█████▎    | 16/30 [00:01<00:00, 14.49it/s]
 60%|██████    | 18/30 [00:01<00:00, 14.32it/s]
 67%|██████▋   | 20/30 [00:01<00:00, 14.49it/s]
 73%|███████▎  | 22/30 [00:01<00:00, 14.62it/s]
 80%|████████  | 24/30 [00:01<00:00, 14.74it/s]
 87%|████████▋ | 26/30 [00:01<00:00, 14.74it/s]
 93%|█████████▎| 28/30 [00:01<00:00, 14.67it/s]
100%|██████████| 30/30 [00:02<00:00, 14.72it/s]
100%|██████████| 30/30 [00:02<00:00, 14.51it/s]
Version Details
Version ID
b8f1324c04d97f36ea5b717eb1bd4fd00bc418f072b4bf167172708e7131d1b1
Version Created
October 7, 2022
Run on Replicate →