Get up and running in under 5 minutes.

    For Node Operators

    Turn your GPU into an inference server in 4 commands:

    1. pip install compressnode
    2. compressnode init --tier 1 --endpoint https://your-ip:8080
    3. compressnode pull llama-3.1-8b-compressed
    4. compressnode start --model llama-3.1-8b-compressed

    For Developers

    Submit an inference job via the API:

    1. POST to /api/jobs/submit with your model choice and input prompt
    2. The API returns the assigned node, estimated cost in ETH, and instructions
    3. Call JobRegistry.submitJob() on-chain with the ETH payment as msg.value
    4. Wait for the node to complete inference - the result is returned via the API

    Requirements

    Before you begin, make sure you have:

    • A GPU with at least 8GB VRAM (RTX 3060 or newer recommended)
    • Python 3.11+ installed on your system
    • A Base-compatible wallet with some ETH for gas fees
    • A stable internet connection with a public endpoint (static IP or domain)

    Wallet Setup

    CompressNode uses Privy for wallet authentication. You can connect with MetaMask, Coinbase Wallet, WalletConnect, or create an embedded wallet via email. All payments and registrations happen on the Base chain - make sure your wallet is configured for Base.