MCP server · Developer tools · official
AmpleRun MCP server
by ample-run · official
Lets your AI find, price, rent and stop cloud GPUs on AmpleRun, within a budget you set.
This is a small helper that connects your AI assistant to AmpleRun, a service where you can rent computers with powerful graphics chips (GPUs) by the hour. It is handy if you need a GPU for a task like training or running an AI model, and you would rather ask your assistant in plain words than click around a website. By default it can only look things up and give you prices, and it cannot spend anything until you turn spending on yourself.
What is an MCP server? The 30-second version
On its own, your AI can only chat. An MCP server is a small helper program that gives your AI a new skill or a connection to an app or service. This one connects your AI to AmpleRun, so it can look up available GPUs, get prices, and, if you allow it, rent a machine for you. Think of it as giving your assistant a phone line to AmpleRun, with a spending limit you control.
What this MCP server does
You ask your AI something like which GPUs are free right now or how much a rental would cost. The AI passes your question to this helper, and the helper talks to AmpleRun's website on your behalf. AmpleRun sends back a list of machines, prices, or a quote, and the AI shows it to you in the chat. If you have turned spending on, the AI can also rent a machine within a budget you set, check on it, and stop it when you are done.
What you can do with it
- Browse ready-made templates for things like PyTorch or an inference server
- Search available GPU machines, cheapest first, and filter by model, memory, price or region
- Get a free 60-second price quote for a template on a specific machine
- Rent a machine with a hard budget cap you set in advance
- Check the state of a rental and get its access details, like an SSH command
- See your USDC and USDT balance and recent spend
- Stop a rental so it stops charging you
Try asking your AI
- “What GPUs are available right now under 2 dollars an hour?”
- “Get me a quote for the PyTorch template on the cheapest offer with at least 24 GB of VRAM.”
- “Rent that machine with a budget of 5 USDC and tell me the rental ID.”
- “Show me my current rentals and stop the one that is still running.”
What it gives back to you
You get answers in plain chat: lists of machines with prices, a quote with the fee split, or a rental ID with its state and access details. Balance checks come back as numbers in USDC and USDT, with spendable, held and recent spend shown separately. When you rent or stop something, the AI tells you what changed and gives you the ID so you can refer to it later.
Before you start
What you need
- Node.js 20 or later if you run it on your own computer
- An AmpleRun API key with the renter scope from Account, Security (only needed for account tools; browsing and pricing work without it)
- A funded USDC or USDT balance on Base if you want to rent
- The AMPLERUN_MCP_ALLOW_SPEND setting turned on if you want the AI to be able to rent or stop
Good to know
If you turn on spending, the AI can rent machines and spend your USDC or USDT balance, so set a budget and use a dedicated wallet holding only what you are willing to spend.
Install it with your AI
Add AmpleRun MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check AmpleRun MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,609 MCP servers here for you.
Who it's for
People who need a GPU now and then, like developers, data scientists or AI tinkerers, and anyone who would rather ask their assistant than use a website.
