Run a Metafog AI node
Offer your compute resources and join the Metafog providers.
Prerequisites
Before you begin, ensure you meet the following requirements:
- Windows 10 workstation - Core i5 or above, 16GB RAM, 50GB free disk
- Docker and Docker Compose installed on your system
- A stable internet connection (wired connection preferred)
- Basic familiarity with terminal/command-line operations
- [Optional] NVIDIA GPU(s) with CUDA support - 8GB VRAM minimum
Metafog eliminates the need for a public or static IP address! Its technology enables it to function seamlessly behind a home router, NAT, or firewall.
Run the Metafog Node
Follow these simple steps to set up your Metafog AI node:
Clone the Metafog Repository
git clone https://github.com/metafogio/metafog-provider.git
cd metafog-provider
Update the Provider Key
- Login to Metafog
- Navigate to Providers > Dashboard in the left menu
- Copy your
METAFOG_PROVIDER_KEY
- Open
docker-compose.yaml
(ordocker-compose.gpu.yaml
if using GPU) and update your provider key:
# docker-compose.yaml (or docker-compose.gpu.yaml if using GPU)
- METAFOG_PROVIDER_KEY = PASTE_YOUR_METAFOG_PROVIDER_KEY
Start Your Node
Choose the appropriate command based on your hardware:
Option 1: Standard Setup (CPU Only)
If you're running without GPU, simply use:
docker compose up -d
Option 2: GPU-Enabled Setup
If you have NVIDIA GPU(s), use this command to start the node with GPU support:
docker compose -f docker-compose.gpu.yaml up -d
This command pulls the required images, initializes the node, and connects it to the Metafog network.
Verify Your Node
Check the logs to confirm your node is running and connected:
# For CPU setup
docker compose logs -f
# For GPU setup
docker compose -f docker-compose.gpu.yaml logs -f
You should see connection messages indicating your node is successfully connected to the Metafog network.