Quickstart

Thunder Compute provides an MCP (Model Context Protocol) server that lets AI coding agents manage GPU instances on your behalf — create, monitor, modify, and tear down instances without leaving your agent workflow.

Prerequisites

A Thunder Compute account — sign up here
A payment method on your account
An AI agent that supports remote MCP servers (Claude Code, Cursor, Windsurf, etc.)

No local installation or API tokens required — authentication is handled via OAuth in your browser.

Setup

Claude Code
Cursor
Windsurf

Run this in Claude Code:

/mcp add --transport http https://api.thundercompute.com:8443/mcp

Then run /mcp to authenticate. A browser window will open for you to log in and authorize access.

Add to .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "thunder-compute": {
      "type": "http",
      "url": "https://api.thundercompute.com:8443/mcp"
    }
  }
}

Add to your MCP configuration:

{
  "mcpServers": {
    "thunder-compute": {
      "serverUrl": "https://api.thundercompute.com:8443/mcp",
      "headers": {
        "Content-Type": "application/json"
      }
    }
  }
}

Try It Out

Once configured, ask your AI agent to:

“Spin up an A100 instance with PyTorch”
“List my running instances”
“Run nvidia-smi on my instance”
“How much have I spent this month?”

The MCP server gives your agent access to instance management, snapshots, SSH keys, port forwarding, billing, and more.

Next Steps

Read the full MCP Server guide for the complete list of available tools, prompts, and troubleshooting
Learn about Prototyping vs Production to choose the right mode for your workload
Explore Technical Specifications for hardware, networking, and storage details

Getting Started

Guides

Reference

API Reference

Prerequisites

Setup

Try It Out

Next Steps

Getting Started

Guides

Reference

API Reference

​Prerequisites

​Setup

​Try It Out

​Next Steps

Prerequisites

Setup

Try It Out

Next Steps