Documentation Index Fetch the complete documentation index at: https://www.thundercompute.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
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
An AI agent that supports remote MCP servers (Claude Code, Cursor, Codex, etc.)
No local installation or API tokens required — authentication is handled via OAuth in your browser.
Setup Claude Code
Codex
Cursor
Windsurf
OpenCode
Smithery
Direct HTTP
Run this in your terminal: claude mcp add --transport http thunder-compute https://www.thundercompute.com/mcp
Then start Claude Code and run /mcp to authenticate. A browser window will open for you to log in and authorize access.
Alternatively, add to ~/.claude.json (global) or .claude.json in your project root: { "mcpServers" : { "thunder-compute" : { "url" : "https://www.thundercompute.com/mcp" } } }
Run this in your terminal: codex mcp add thunder-compute --url https://www.thundercompute.com/mcp
Codex will prompt you to authenticate via OAuth when you first use a Thunder Compute tool. Add to .cursor/mcp.json in your project root (or ~/.cursor/mcp.json for global access): { "mcpServers" : { "thunder-compute" : { "type" : "http" , "url" : "https://www.thundercompute.com/mcp" } } }
Add to your MCP configuration: { "mcpServers" : { "thunder-compute" : { "serverUrl" : "https://www.thundercompute.com/mcp" , "headers" : { "Content-Type" : "application/json" } } } }
Add the MCP server
Run the interactive setup: When prompted:
Server name: thunder-computeServer type: RemoteURL: https://www.thundercompute.com/mcpRequires OAuth: YesPre-registered client ID: No
Authenticate
opencode mcp auth thunder-compute
A browser window will open for you to log in and authorize access.
Start coding
The Thunder Compute tools are now available in your session.
Alternatively, add to ~/.config/opencode/opencode.json: { "$schema" : "https://opencode.ai/config.json" , "mcp" : { "thunder-compute" : { "type" : "remote" , "url" : "https://www.thundercompute.com/mcp" , "oauth" : {} } } }
Then run opencode mcp auth thunder-compute to authenticate. If you use an MCP client that supports Smithery , you can install directly: npx @smithery/cli install @thunder-compute/thunder-compute
Or browse the Thunder Compute listing on Smithery and click Install for your client. For custom integrations, the MCP server uses Streamable HTTP transport at a single endpoint. Authentication is via OAuth 2.0 with standard MCP discovery. Endpoint: https://www.thundercompute.com/mcpcurl -X POST https://www.thundercompute.com/mcp \ -H "Authorization: Bearer <access-token>" \ -H "Content-Type: application/json" \ -d '{ "jsonrpc": "2.0", "method": "initialize", "params": { "protocolVersion": "2025-03-26", "capabilities": {}, "clientInfo": { "name": "my-agent", "version": "1.0.0" } }, "id": 1 }'
Authentication No API tokens or environment variables needed. When you first connect, a browser window opens for you to log in with your Thunder Compute account and authorize access. Tokens refresh automatically, so you only authenticate once per session.
Instance Management Tool Description list_instancesList all GPU instances with status, IP, and configuration create_instanceCreate a new GPU instance (specify GPU type, template, mode, etc.) delete_instanceDelete an instance (irreversible) modify_instanceChange instance config (GPU type, vCPUs, disk, mode) run_commandExecute a shell command on a running instance and return stdout, stderr, and exit code
Tool Description get_specsGet available GPU specs (VRAM, vCPU options, storage ranges) get_availabilityGet current GPU availability status for each spec get_pricingGet current per-hour GPU pricing list_templatesList available OS templates (Ubuntu, PyTorch, etc.)
Snapshots Tool Description list_snapshotsList all instance snapshots create_snapshotCreate a snapshot of an instance delete_snapshotDelete a snapshot (irreversible)
SSH Keys Tool Description list_ssh_keysList SSH keys in your organization create_ssh_keyAdd an SSH public key to your organization delete_ssh_keyDelete an SSH key add_ssh_key_to_instanceAdd an SSH public key to a running instance’s authorized_keys
Port Forwarding Tool Description list_portsList all instances with their forwarded ports forward_portForward HTTP ports on an instance delete_portRemove forwarded ports from an instance
Connectivity Tool Description get_ssh_commandGet the SSH command to connect to an instance get_scp_commandGet the SCP command to copy files to/from an instance
Billing & Usage Tool Description get_meter_dataGet GPU usage metrics for a time period (hourly, daily, weekly, or monthly) get_upcoming_invoiceGet estimated charges for the current billing period get_invoice_historyGet historical invoices for your organization get_subscriptionGet subscription details including plan, status, and payment info
API Tokens Tool Description list_tokensList all named API tokens for your organization create_tokenCreate a new named API token delete_tokenDelete a named API token
Prompts The MCP server includes built-in prompts that guide your agent through common multi-step workflows:
Prompt Description create-dev-instanceSet up a GPU development instance with sensible defaults deploy-modelDeploy an ML model (supports Ollama, vLLM, and Transformers) check-costsReview current GPU usage and costs snapshot-and-teardownSave instance state and clean up setup-comfyuiSpin up a GPU instance with ComfyUI for AI image generation setup-jupyterLaunch a Jupyter Lab environment on a GPU instance fine-tune-modelSet up a GPU instance for fine-tuning with LoRA or full fine-tuning benchmark-gpuRun a quick GPU benchmark on an instance to verify performance
Example Usage Once configured, you can ask your AI agent things like:
“Spin up an A100 instance with PyTorch”
“What GPU types are available and how much do they cost?”
“Which GPUs are available right now?”
“List my running instances”
“Run nvidia-smi on my instance”
“Delete instance inst-abc123”
“Forward port 8080 on my instance”
“Create a snapshot of my instance before I make changes”
“Deploy Llama 3 on a GPU”
“How much have I spent this month?”
“Show my invoice history”
“Create an API token for my CI pipeline”
Troubleshooting Authentication fails or browser doesn’t open : Run /mcp in Claude Code to manually trigger authentication. Make sure you’re logged in to your Thunder Compute account in the browser.Browser callback fails after clicking Allow (common when running your AI agent inside a remote VM, container, or SSH session): if the browser shows a “can’t connect” / “site can’t be reached” page after consent, copy the full URL from the browser’s address bar (it looks like http://localhost:<port>/callback?code=...&state=...) and paste it into your AI agent where it prompts you to finish authentication. The agent will extract the code from the URL and complete the token exchange itself.“Protected resource does not match” error : The URL in your MCP config must match the server’s configured resource URL exactly. Ensure you’re using https://www.thundercompute.com/mcp.“token has invalid issuer” error : This is a server-side configuration issue. The MCP authentication client must be configured with the correct Stytch Connected Apps domain.Tools not appearing : Restart your AI agent after changing MCP configuration. Most agents only read MCP config on startup.MCP Directories Thunder Compute is listed on major MCP directories for easy discovery:
Smithery MCP Registry Glama PulseMCP
If your MCP client supports browsing directories, search for “Thunder Compute” to find and install the server directly. 
