stability-ai/stable-diffusion-3.5-medium 🔢🖼️📝❓ → 🖼️

⭐ Official ▶️ 76.6K runs 📅 Oct 2024 ⚙️ Cog 0.16.7 📄 Paper ⚖️ License
image-to-image text-to-image typography

About

2.5 billion parameter image model with improved MMDiT-X architecture

Example Output

Prompt:

"a captivating anime-style illustration of a woman in a white astronaut suit. She has long, dark wavy hair. Surrounding the astronaut are vibrant orange flowers with yellow centers. The background itself is a mesmerizing night sky filled with countless stars"

Output

Example output

Performance Metrics

4.39s Prediction Time
4.40s Total Time
1 Images
All Input Parameters
{
  "cfg": 5,
  "steps": 40,
  "prompt": "a captivating anime-style illustration of a woman in a white astronaut suit. She has long, dark wavy hair. Surrounding the astronaut are vibrant orange flowers with yellow centers. The background itself is a mesmerizing night sky filled with countless stars",
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "output_quality": 90,
  "prompt_strength": 0.85
}
Input Parameters
cfg Type: numberDefault: 5Range: 1 - 10
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.
prompt (required) Type: string
Text prompt for image generation
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
negative_prompt Type: string
What you do not want to see in the image
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: stringFormat: uri

Example Execution Logs
Seed set to: 3551813581
Running workflow
got prompt
Executing node 294, title: KSampler, class type: KSampler
0%|          | 0/40 [00:00<?, ?it/s]
5%|▌         | 2/40 [00:00<00:03, 10.76it/s]
10%|█         | 4/40 [00:00<00:03, 10.78it/s]
15%|█▌        | 6/40 [00:00<00:03, 10.76it/s]
20%|██        | 8/40 [00:00<00:02, 10.77it/s]
25%|██▌       | 10/40 [00:00<00:02, 10.78it/s]
30%|███       | 12/40 [00:01<00:02, 10.78it/s]
35%|███▌      | 14/40 [00:01<00:02, 10.78it/s]
40%|████      | 16/40 [00:01<00:02, 10.78it/s]
45%|████▌     | 18/40 [00:01<00:02, 10.78it/s]
50%|█████     | 20/40 [00:01<00:01, 10.79it/s]
55%|█████▌    | 22/40 [00:02<00:01, 10.77it/s]
60%|██████    | 24/40 [00:02<00:01, 10.76it/s]
65%|██████▌   | 26/40 [00:02<00:01, 10.77it/s]
70%|███████   | 28/40 [00:02<00:01, 10.76it/s]
75%|███████▌  | 30/40 [00:02<00:00, 10.77it/s]
80%|████████  | 32/40 [00:02<00:00, 10.76it/s]
85%|████████▌ | 34/40 [00:03<00:00, 10.76it/s]
90%|█████████ | 36/40 [00:03<00:00, 10.77it/s]
95%|█████████▌| 38/40 [00:03<00:00, 10.78it/s]
100%|██████████| 40/40 [00:03<00:00, 10.79it/s]
100%|██████████| 40/40 [00:03<00:00, 10.77it/s]
Executing node 8, title: VAE Decode, class type: VAEDecode
Executing node 309, title: Save Image, class type: SaveImage
Prompt executed in 4.03 seconds
outputs:  {'309': {'images': [{'filename': 'R8_sd3.5L_00001_.png', 'subfolder': '', 'type': 'output'}]}}
====================================
R8_sd3.5L_00001_.png
Version Details
Version ID
e069811fe9b5815a6db8874c3054c15be99c8019d07ee5ad72aa63cea3cfffe3
Version Created
September 26, 2025
Run on Replicate →