Back to articles
AI & Cloud

Guide to Running Stable Diffusion ComfyUI on VPS GPU for 24/7 AI Art Rendering with Optimized Costs

A detailed guide on deploying Stable Diffusion ComfyUI on a VPS GPU to create AI art 24/7 with optimized costs, including system configuration, installation, performance tuning, and...

DPTCloud Engineering4 min read
Guide to Running Stable Diffusion ComfyUI on VPS GPU for 24/7 AI Art Rendering with Optimized Costs

Stable Diffusion is one of the most popular AI image generation models, enabling users to create high-quality images from text descriptions. To effectively leverage this technology, deploying Stable Diffusion ComfyUI on a VPS GPU is an ideal choice for those wanting to generate AI art 24/7 with optimized costs. This guide walks you through setting up a ComfyUI environment on a VPS GPU, optimizing performance, and ensuring stable, continuous operation.

System Requirements and Preparations

To run Stable Diffusion ComfyUI efficiently, you need a VPS with a GPU that supports CUDA (such as NVIDIA T4, A10, A100, or equivalent). Providers like High-RAM Cloud VPS for Self-Hosted LLMs & API Gateways offer GPU-enabled VPS packages suitable for this use case. Before starting, verify the following:

  • GPU acceleration is enabled on your VPS.
  • Docker and Docker Compose are installed.
  • Git is installed to clone the source code.
  • You have sudo or root access for installation.

Setting Up ComfyUI on VPS GPU

The first step is installing ComfyUI. You can clone the GitHub repository and run it using Docker to ensure a stable environment.

git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI
# Create docker-compose.yml with GPU configuration
cat > docker-compose.yml << EOF
version: '3.8'
services:
  comfyui:
    image: ghcr.io/comfyanonymous/ComfyUI:latest
    container_name: comfyui
    ports:
      - "8188:8188"
    volumes:
      - ./models:/root/ComfyUI/models
      - ./custom_nodes:/root/ComfyUI/custom_nodes
    devices:
      - /dev/dri:/dev/dri
    runtime: nvidia
    environment:
      - NVIDIA_VISIBLE_DEVICES=all
    restart: unless-stopped
EOF
docker-compose up -d

This command downloads and runs ComfyUI on your VPS GPU, exposing port 8188 for web access. You can access the interface at http://[VPS_IP]:8188 to begin generating images.

Optimizing Performance and Costs

To reduce costs and improve performance, apply the following strategies:

  • Use smaller GPUs like NVIDIA T4 if high performance isn't required.
  • Run ComfyUI only when needed by setting restart: unless-stopped and stopping the container when not in use.
  • LIMIT the number of models loaded—only download and use the models you need.
  • Optimize Docker Compose by reducing memory and CPU allocations for the container.

Verification and Troubleshooting

After installation, verify ComfyUI is running correctly by checking the logs:

docker-compose logs -f comfyui

If GPU-related errors occur, verify the configuration using:

nvidia-smi

If nvidia-smi does not display GPU devices, recheck the driver installation and Docker runtime configuration.

References

DPTCloud Tier III Infrastructure

High-RAM Cloud VPS Plans Optimized for AI, Databases & Cache

Enterprise RAID 10 NVMe SSDs, CMC & FPT District 9 uplinks with Anycast Anti-DDoS and 99.99% SLA.

View 22 VPS Plans
ECO · vn-02Recommended

VPS-4E

297.000 ₫ / mo

10% VAT included

  • CPU: 4 vCPU
  • RAM: 8 GB
  • SSD: 50 GB NVMe
ECO · vn-02

VPS-5E

614.900 ₫ / mo

10% VAT included

  • CPU: 8 vCPU
  • RAM: 16 GB
  • SSD: 100 GB NVMe
ADVANCED · vn-03Recommended

VPS-4T

548.900 ₫ / mo

10% VAT included

  • CPU: 4 vCPU
  • RAM: 8 GB
  • SSD: 50 GB NVMe

Related Articles

View all
ZaloZalo SupportDiscord0384 753 478