Stable Diffusion WebUI

Stable Diffusion WebUI (commonly referring to AUTOMATIC1111’s stable-diffusion-webui project) is the most popular graphical interface for Stable Diffusion. It offers rich features including text-to-image (txt2img), image-to-image (img2img), upscaling, ControlNet, and more. It is accessible through a browser and well-suited for deployment on GPU servers for team use.

System Requirements

• CentOS Stream 9 & 10 / AlmaLinux 9 & 10 / Rocky Linux 9 & 10
• Python 3.10 or 3.11 (EL 10 defaults to Python 3.12, but SD WebUI recommends 3.10/3.11, so an older version must be installed manually)
• Git
• NVIDIA GPU with at least 6 GB VRAM (8 GB or more recommended)
• NVIDIA drivers and CUDA installed (see NVIDIA Drivers and CUDA)
• At least 20 GB disk space (model files are large)

Pre-Installation

Install System Dependencies

sudo dnf install -y git python3.11 python3.11-devel python3.11-pip \
  gcc gcc-c++ make wget \
  libffi-devel openssl-devel bzip2-devel \
  mesa-libGL

Note

Stable Diffusion WebUI recommends Python 3.10 or 3.11. If these versions are not available by default on your system, you may need to compile from source or install from a third-party repository.

Verify GPU Environment

nvidia-smi

Make sure the output shows your GPU model and driver version.

Create a Dedicated User (Optional but Recommended)

sudo useradd -m -s /bin/bash sduser
sudo su - sduser

Installing Stable Diffusion WebUI

Clone the Repository

cd /opt
sudo mkdir -p stable-diffusion
sudo chown $(whoami):$(whoami) stable-diffusion
cd stable-diffusion

git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
cd stable-diffusion-webui

Download Models

You need at least one Stable Diffusion model checkpoint file. Download a model from Hugging Face or Civitai and place it in the models/Stable-diffusion/ directory:

# Download Stable Diffusion v1.5 (approximately 4 GB)
wget -P models/Stable-diffusion/ \
  https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors

# Or download SDXL Base (approximately 6.5 GB)
wget -P models/Stable-diffusion/ \
  https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors

First Launch

The WebUI launch script will automatically create a virtual environment and install all Python dependencies:

cd /opt/stable-diffusion/stable-diffusion-webui

# Launch using Python 3.11
python3.11 launch.py --listen --xformers

Parameter descriptions:

• --listen: Listen on all network interfaces (default is 127.0.0.1 only), allowing remote access
• --xformers: Enable xformers acceleration, reducing VRAM usage and speeding up generation

The first run will download and install a large number of dependencies, which may take 10-20 minutes depending on network speed.

Once installation is complete, you will see output similar to:

Running on local URL:  http://0.0.0.0:7860

Common Startup Parameters

python3.11 launch.py \
  --listen \
  --port 7860 \
  --xformers \
  --enable-insecure-extension-access \
  --api \
  --medvram

Parameter	Description
--listen	Listen on all interfaces
--port	Specify port number
--xformers	Enable xformers memory optimization
--api	Enable REST API endpoint
--medvram	Medium VRAM optimization (recommended for 6-8 GB VRAM)
--lowvram	Low VRAM optimization (for 4-6 GB VRAM)
--enable-insecure-extension-access	Allow installing extensions from the WebUI
--gradio-auth user:pass	Set access username and password
--no-half-vae	Try this parameter if generated images are black

Basic Usage

Accessing the WebUI

Open http://your-server-ip:7860 in your browser.

Text-to-Image (txt2img)

1. Select the loaded model at the top
2. Enter English prompt words in the Prompt box describing the image you want to generate
3. Enter elements you do not want in the Negative Prompt box
4. Adjust parameters:
   • Sampling method: DPM++ 2M Karras or Euler a recommended
   • Sampling steps: 20-30 steps
   • Width/Height: 512x512 recommended for SD 1.5; 1024x1024 recommended for SDXL
   • CFG Scale: 7-12
   • Seed: -1 for random
5. Click Generate

Image-to-Image (img2img)

Switch to the img2img tab, upload a reference image, fill in the prompt, adjust the Denoising strength (0.3-0.7), and click Generate.

API Calls

When launched with the --api parameter, you can generate images via the HTTP API:

curl -X POST http://localhost:7860/sdapi/v1/txt2img \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a beautiful sunset over mountains, photorealistic",
    "negative_prompt": "blurry, low quality",
    "steps": 25,
    "width": 512,
    "height": 512,
    "cfg_scale": 7
  }'

API documentation is available at http://your-server-ip:7860/docs.

systemd Service

Create a systemd service for management and auto-start:

sudo tee /etc/systemd/system/stable-diffusion.service > /dev/null <<'EOF'
[Unit]
Description=Stable Diffusion WebUI
After=network-online.target
Wants=network-online.target

[Service]
Type=simple
User=sduser
Group=sduser
WorkingDirectory=/opt/stable-diffusion/stable-diffusion-webui
ExecStart=/usr/bin/python3.11 launch.py --listen --port 7860 --xformers --api --enable-insecure-extension-access
Restart=on-failure
RestartSec=10
Environment="HOME=/home/sduser"

# Resource limits
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

Enable and start the service:

sudo systemctl daemon-reload
sudo systemctl enable --now stable-diffusion

Manage the service:

# Check status
sudo systemctl status stable-diffusion

# View logs
sudo journalctl -u stable-diffusion -f

# Restart
sudo systemctl restart stable-diffusion

Firewall Configuration

sudo firewall-cmd --permanent --add-port=7860/tcp
sudo firewall-cmd --reload

Security Recommendations

Production Environment Security

Stable Diffusion WebUI has a relatively simple built-in authentication mechanism. If deploying on the public internet, it is strongly recommended to:

• Set an access password using --gradio-auth username:password
• Use an Nginx reverse proxy with HTTPS configured
• Restrict source IPs with firewall rules

Example Nginx reverse proxy configuration:

sudo tee /etc/nginx/conf.d/sd-webui.conf > /dev/null <<'EOF'
server {
    listen 443 ssl;
    server_name sd.example.com;

    ssl_certificate /etc/letsencrypt/live/sd.example.com/fullchain.pem;
    ssl_certificate_key /etc/letsencrypt/live/sd.example.com/privkey.pem;

    client_max_body_size 100M;

    location / {
        proxy_pass http://127.0.0.1:7860;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade";
        proxy_read_timeout 300s;
    }
}
EOF

FAQ

Startup Error: No module named pip

python3.11 -m ensurepip --upgrade

Black or Solid-Color Images

Try adding the --no-half or --no-half-vae startup parameters:

python3.11 launch.py --listen --xformers --no-half-vae

Out of Memory

# Enable VRAM optimization
python3.11 launch.py --listen --xformers --medvram

# More aggressive VRAM optimization
python3.11 launch.py --listen --xformers --lowvram

# Reduce generation size
# In the WebUI, reduce Width/Height to 512x512

Slow pip Dependency Installation

Set a pip mirror:

mkdir -p ~/.config/pip
cat > ~/.config/pip/pip.conf <<'EOF'
[global]
index-url = https://mirrors.aliyun.com/pypi/simple/
trusted-host = mirrors.aliyun.com
EOF

Updating the WebUI

cd /opt/stable-diffusion/stable-diffusion-webui
git pull
# Simply restart; dependencies will be updated automatically