Quick start

Give your AI agent GPU access in under 5 minutes.

1. Sign up

Create an account at app.volspan.dev. Every new account gets $5 in free credit loaded into your Vault automatically. No credit card required.

2. Create an API key

Go to Settings → API Keys and create a new key. Name it something descriptive like “agent”. Copy the key immediately — it won't be shown again.

vc_live_aBcDeFgHiJkLmNoPqRsTuVwXyZ012345

3. Configure MCP

Add the Volspan MCP server to your client config. For Claude Desktop, edit claude_desktop_config.json. For Cursor, edit .cursor/mcp.json.

{
  "mcpServers": {
    "volspan": {
      "command": "npx",
      "args": ["-y", "@volspan/mcp-server"],
      "env": {
        "VOLSPAN_API_KEY": "vc_live_your_key_here"
      }
    }
  }
}

4. Run code on a GPU

Ask your agent to run Python on a GPU. It will call the run_on_gpu tool automatically.

You

Run a PyTorch benchmark on an A100. Print the TFLOPS.

Agent

I'll run a matrix multiplication benchmark on an A100 GPU for you.

Calling run_on_gpu with gpu="A100" ...

Next steps

Need help? Reach out at [email protected]