stability-ai/stable-diffusion 🔢❓📝 → 🖼️

▶️ 110.8M runs 📅 Aug 2022 ⚙️ Cog 0.7.2 🔗 GitHub 📄 Paper ⚖️ License
text-to-image

About

A latent text-to-image diffusion model capable of generating photo-realistic images given any text input

Example Output

Prompt:

"an astronaut riding a horse on mars, hd, dramatic lighting"

Output

Example output

Performance Metrics

1.89s Prediction Time
2.04s Total Time
All Input Parameters 
{
  "prompt": "an astronaut riding a horse on mars, hd, dramatic lighting",
  "scheduler": "K_EULER",
  "num_outputs": 1,
  "guidance_scale": 7.5,
  "image_dimensions": "512x512",
  "num_inference_steps": 50
}
Input Parameters
seed Type: integer
Random seed. Leave blank to randomize the seed
width Default: 768
Width of generated image in pixels. Needs to be a multiple of 64
height Default: 768
Height of generated image in pixels. Needs to be a multiple of 64
prompt Type: stringDefault: a vision of paradise. unreal engine
Input prompt
scheduler Default: DPMSolverMultistep
Choose a scheduler.
num_outputs Type: integerDefault: 1Range: 1 - 4
Number of images to generate.
guidance_scale Type: numberDefault: 7.5Range: 1 - 20
Scale for classifier-free guidance
negative_prompt Type: string
Specify things to not see in the output
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: 52443
input_shape: torch.Size([1, 77])
  0%|          | 0/50 [00:00<?, ?it/s]
 10%|█         | 5/50 [00:00<00:01, 41.06it/s]
 20%|██        | 10/50 [00:00<00:00, 41.32it/s]
 30%|███       | 15/50 [00:00<00:00, 43.77it/s]
 42%|████▏     | 21/50 [00:00<00:00, 46.80it/s]
 54%|█████▍    | 27/50 [00:00<00:00, 48.53it/s]
 66%|██████▌   | 33/50 [00:00<00:00, 49.63it/s]
 78%|███████▊  | 39/50 [00:00<00:00, 50.45it/s]
 90%|█████████ | 45/50 [00:00<00:00, 50.90it/s]
100%|██████████| 50/50 [00:01<00:00, 48.51it/s]
Version Details
Version ID
ac732df83cea7fff18b8472768c88ad041fa750ff7682a21affe81863cbe77e4
Version Created
July 10, 2023
Run on Replicate →