How to Run Stable Diffusion Locally

ComfyUI is a node-based interface that gives you complete control over the generation process.

# Clone ComfyUI
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI

# Install dependencies
pip install -r requirements.txt

💡 Requires Python 3.10+ and NVIDIA GPU with CUDA.

Download Stable Diffusion checkpoints and place them in the models folder.

# Download SDXL (recommended)
# From: https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0
# Place in: ComfyUI/models/checkpoints/

# Or use wget:
wget -P models/checkpoints/ https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors

Start the ComfyUI server and access it in your browser.

python main.py

# Open http://127.0.0.1:8188 in your browser

💡 First launch compiles CUDA kernels (takes a few minutes).

Load the default workflow and enter a prompt to generate your first image!

Prompt: "a photo of an astronaut riding a horse on mars, highly detailed, 8k"

Settings: Steps: 20, CFG: 7, Sampler: euler_ancestral