Earn $GITBLOCK

Run a Node.
Get Free API Access.

Contribute your GPU to the GitBlock network. Serve AI models, earn $GITBLOCK rewards, and get unlimited free API access for yourself.

Start Your Node
Why Run a Node

Earn While You Compute

Your GPU powers the decentralized AI network. You get paid in $GITBLOCK and get free API access.

Earn $GITBLOCK

Get paid per request served. Reputation multiplier boosts earnings up to 2x.

Free API Key

Node operators get unlimited free inference. Your key works everywhere on GitBlock.

Reputation Score

Build reputation over time. Higher reputation = higher rewards + priority routing.

Any GPU Works

Run on anything from a gaming GPU to a data center rig. Support for CPU inference too.

Decentralized

No central authority. You choose which models to serve. Full control over your node.

Early Mover Bonus

Early node operators get bonus rewards. Network is new — maximum earning potential now.

Tutorial

Setup Your Node in 3 Steps

From zero to serving models in under 5 minutes. Works on Windows, Mac, and Linux.

1

Install GitBlock

One pip command. Python 3.10+ required. GPU drivers recommended but optional.

2

Configure Your Node

Set your wallet address, pick models to serve. Single config command.

3

Start Serving

One command starts your node. It registers with the network and begins earning.

Step 1: Install

Bash # Install the GitBlock node package pip install gitblock[node] # Verify installation gitblock --version # → GitBlock v0.2.0

Step 2: Configure

Bash # Configure your node with wallet + models gitblock-node \ --wallet 0xYourWalletAddress \ --model llama-3.3-70b \ --model mistral-7b \ --model deepseek-coder-33b # Supported models (80+ available): # llama-3.3-70b, mistral-7b, deepseek-coder-33b, # qwen-2.5-72b, gemma-4-9b, phi-3-14b, # codellama-34b, yi-1.5-34b, and more...

Step 3: Start Serving

Bash # Start your node python -m node # Your node is now live! # → Registered on the GitBlock network # → Serving: llama-3.3-70b, mistral-7b, deepseek-coder-33b # → Earning $GITBLOCK per request # → Health check: http://localhost:8080/health
API Access

Get Your Free API Key

Once your node is running, your wallet address becomes your API key. Use it anywhere on GitBlock.

Python from gitblock import GitBlock # Your wallet IS your API key as a node operator client = GitBlock(api_key="gb_node_" + your_wallet_address[:8]) response = client.chat( model="llama-3.3-70b", messages=[{"role": "user", "content": "Hello!"}] ) print(response.choices[0].message.content)
cURL curl -X POST https://api.gitblock.io/v1/chat/completions \ -H "Authorization: Bearer gb_node_YOURKEY" \ -H "Content-Type: application/json" \ -d '{"model":"llama-3.3-70b","messages":[{"role":"user","content":"Hello!"}]}'
FAQ

Common Questions

Any GPU with 8GB+ VRAM for 7B models, 24GB+ for 70B models. CPU-only inference is supported but slower. NVIDIA GPUs work best (CUDA). AMD and Apple Silicon also supported via llama.cpp.
Earnings depend on your GPU, uptime, and reputation score. A node running 24/7 with a 70B model can earn 100-500 $GITBLOCK per day. Reputation multiplier can double this. Early operators get bonus rates.
Yes. Node operators get unlimited free inference as a perk of contributing to the network. Your wallet address becomes your API key. Use it in the Console, with the SDK, or via REST API.
Yes. Each node needs a unique port and can serve different models. Earnings stack per node. Many operators run a fleet of nodes across different machines.