How to beginner · 3 min read

How to run Stable Diffusion locally

Q: How to run Stable Diffusion locally

Run Stable Diffusion locally by installing the diffusers Python library and its dependencies, then load a pre-trained model like runwayml/stable-diffusion-v1-5 with StableDiffusionPipeline. Use a GPU-enabled environment for best performance and generate images by passing prompts to the pipeline.

Quick answer

Run Stable Diffusion locally by installing the diffusers Python library and its dependencies, then load a pre-trained model like runwayml/stable-diffusion-v1-5 with StableDiffusionPipeline. Use a GPU-enabled environment for best performance and generate images by passing prompts to the pipeline.

PREREQUISITES

Python 3.8+
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117 (for CUDA GPU)
pip install diffusers transformers scipy ftfy
A CUDA-compatible GPU (recommended) or CPU fallback

Setup

Install the required Python packages and ensure you have a CUDA-enabled GPU for optimal performance. Use the official PyTorch installation command for your system and then install the diffusers library along with dependencies.

bash

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
pip install diffusers transformers scipy ftfy

Step by step

Use the following Python code to load the Stable Diffusion model and generate an image from a text prompt. This example uses the runwayml/stable-diffusion-v1-5 model and outputs the generated image to a file.

python

import torch
from diffusers import StableDiffusionPipeline

# Load the pipeline with the model
pipe = StableDiffusionPipeline.from_pretrained(
    "runwayml/stable-diffusion-v1-5",
    torch_dtype=torch.float16
)
pipe = pipe.to("cuda")  # Use GPU

# Generate an image from a prompt
prompt = "A fantasy landscape with mountains and a river"
image = pipe(prompt).images[0]

# Save the image
image.save("output.png")
print("Image saved as output.png")

output

Image saved as output.png

Common variations

For CPU-only systems, remove torch_dtype=torch.float16 and pipe.to("cuda"), but expect slower generation.
Use other models like stabilityai/stable-diffusion-xl-base-1.0 by changing the model name in from_pretrained.
Enable mixed precision or use torch_dtype=torch.float32 if you encounter precision issues.

Troubleshooting

If you get CUDA out of memory errors, reduce the batch size or use a smaller model.
Ensure your GPU drivers and CUDA toolkit are up to date.
If diffusers fails to load the model, verify your internet connection or download the model manually from Hugging Face.

✅

Key Takeaways

Use the diffusers library with a CUDA-enabled GPU for efficient local Stable Diffusion inference.
Install PyTorch with the correct CUDA version before installing diffusers and dependencies.
Adjust model and precision settings based on your hardware capabilities for best results.

Verified 2026-04 · runwayml/stable-diffusion-v1-5, stabilityai/stable-diffusion-xl-base-1.0

Verify ↗