stability-ai/stable-diffusion-3 🔢🖼️📝❓ → 🖼️

⭐ Official ▶️ 1.8M runs 📅 Jun 2024 ⚙️ Cog 0.9.7 🔗 GitHub 📄 Paper ⚖️ License

About

A text-to-image model with greatly improved performance in image quality, typography, complex prompt understanding, and resource-efficiency

Example Output

Prompt:

"a photo of vibrant artistic graffiti on a wall saying "SD3 medium""

Output

Performance Metrics

2.78s Prediction Time

2.32s Total Time

1 Images

All Input Parameters

{
  "cfg": 3.5,
  "prompt": "a photo of vibrant artistic graffiti on a wall saying \"SD3 medium\"",
  "aspect_ratio": "3:2",
  "output_format": "webp",
  "output_quality": 90,
  "negative_prompt": ""
}

Input Parameters

cfg Type: numberDefault: 3.5Range: 0 - 20: The guidance scale tells the model how similar the output should be to the prompt.
seed Type: integer: Set a seed for reproducibility. Random by default.
image Type: string: Input image for image to image mode. The aspect ratio of your output will match this image.
steps Type: integerDefault: 28Range: 1 - 28: Number of steps to run the sampler for.
prompt Type: stringDefault:
aspect_ratio Default: 1:1: The aspect ratio of your output image. This value is ignored if you are using an input image.
output_format Default: webp: Format of the output images
output_quality Type: integerDefault: 90Range: 0 - 100: Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.
negative_prompt Type: stringDefault:: Negative prompts do not really work in SD3. Using a negative prompt will change your output in unpredictable ways.
prompt_strength Type: numberDefault: 0.85Range: 0 - 1: Prompt strength (or denoising strength) when using image to image. 1.0 corresponds to full destruction of information in image.

Output Schema

Output

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

Example Execution Logs

Seed set to: 448433150
Running workflow
got prompt
Executing node 6, title: CLIP Text Encode (Prompt), class type: CLIPTextEncode
Executing node 71, title: CLIP Text Encode (Negative Prompt), class type: CLIPTextEncode
Executing node 67, title: ConditioningZeroOut, class type: ConditioningZeroOut
Executing node 68, title: ConditioningSetTimestepRange, class type: ConditioningSetTimestepRange
Executing node 70, title: ConditioningSetTimestepRange, class type: ConditioningSetTimestepRange
Executing node 69, title: Conditioning (Combine), class type: ConditioningCombine
Executing node 135, title: EmptySD3LatentImage, class type: EmptySD3LatentImage
Executing node 271, title: KSampler, class type: KSampler
  0%|          | 0/28 [00:00<?, ?it/s]
  7%|▋         | 2/28 [00:00<00:01, 14.59it/s]
 14%|█▍        | 4/28 [00:00<00:01, 15.14it/s]
 21%|██▏       | 6/28 [00:00<00:01, 15.31it/s]
 29%|██▊       | 8/28 [00:00<00:01, 15.40it/s]
 36%|███▌      | 10/28 [00:00<00:01, 15.44it/s]
 43%|████▎     | 12/28 [00:00<00:01, 15.47it/s]
 50%|█████     | 14/28 [00:00<00:00, 15.48it/s]
 57%|█████▋    | 16/28 [00:01<00:00, 15.50it/s]
 64%|██████▍   | 18/28 [00:01<00:00, 15.51it/s]
 71%|███████▏  | 20/28 [00:01<00:00, 15.51it/s]
 79%|███████▊  | 22/28 [00:01<00:00, 15.52it/s]
 86%|████████▌ | 24/28 [00:01<00:00, 15.52it/s]
 93%|█████████▎| 26/28 [00:01<00:00, 15.51it/s]
100%|██████████| 28/28 [00:01<00:00, 15.49it/s]
100%|██████████| 28/28 [00:01<00:00, 15.44it/s]
Executing node 231, title: VAE Decode, class type: VAEDecode
Executing node 273, title: Save Image, class type: SaveImage
Prompt executed in 2.28 seconds
outputs:  {'273': {'images': [{'filename': 'R8_SD3_00001_.png', 'subfolder': '', 'type': 'output'}]}}
====================================
R8_SD3_00001_.png

Version Details

Version ID: 527d2a6296facb8e47ba1eaf17f142c240c19a30894f437feee9b91cc29d8e4f
Version Created: July 16, 2024

Run on Replicate →