We and selected third parties use cookies or similar technologies for technical purposes and, with your consent, to analyze how the site is used.

You can freely give, refuse or withdraw your consent at any time.

CloudGPU

MCP server by mrhy7176

Rent a per-minute GPU from your MCP client, run a job, download the results and shut the GPU down.

Runs locally (Runs on your machine: its page has the install command)Runs with npxNeeds an API key

Install CloudGPU

Runs on your machine. Your client starts it with this command:

npx -y cloudgpu-mcp

Needs Node.js on your computer. The setup steps below show where the command goes in each client.

  • CLOUDGPU_API_KEY secret CloudGPU API key (starts with cgw-sk-). Create one at https://cloudgpu.app/api

Share this server

Opens this page, with the install command and setup steps.

mcp.tc/i/cloudgpu

About

Connects an AI assistant to CloudGPU so it can check live GPU prices, get a quote, rent a GPU by the minute, upload code and data, run commands, tail logs, download results and destroy the instance. It also offers job tools that run on CloudGPU's servers, and lists models on CloudGPU's OpenAI-compatible API.

Runs locally over stdio with npx -y cloudgpu-mcp and needs Node.js 18 or newer. Access needs a CloudGPU API key in CLOUDGPU_API_KEY and a prepaid balance. The assistant can also request a key through a browser approval step. Budgets, time limits and idle limits are enforced server-side. Remote access is also available over Streamable HTTP at https://cloudgpu.app/mcp; the mcp.tc/i/cloudgpu page is a directory listing, not a server endpoint. Local stdio with npx -y cloudgpu-mcp remains supported.

What you can do

  • List GPU models with live prices and availability
  • Get a cost quote before renting
  • Create and stop GPU instances with budget and time limits
  • Run commands and background jobs on the instance
  • Upload and download files over SFTP
  • Run one-off jobs that start and destroy a GPU on the server side
  • List models on the OpenAI-compatible API

Tools

CloudGPU lists its tools only to clients with a key, so we can’t show them here.

Example prompts

  • List the GPUs available and estimate the cost of a 24 GB GPU for one hour.

  • Fine-tune a 7B model with LoRA on data/train.jsonl and save the adapter to ./outputs.

  • Check the status and logs of my running GPU instance.

  • Stop the GPU instance after downloading the results.

Set up

Every client starts CloudGPU on your machine with the same command. Pick yours; the page remembers your choice.

Note from the owner: Remote: connect to https://cloudgpu.app/mcp using Streamable HTTP. mcp.tc/i/cloudgpu is a directory page, not the server endpoint. Local stdio remains supported with npx -y cloudgpu-mcp.

Claude Code

  1. Run this in a terminal, in your project folder:
claude mcp add --transport stdio cloudgpu --env "CLOUDGPU_API_KEY=<YOUR_CLOUDGPU_API_KEY>" -- npx -y cloudgpu-mcp
  1. Start Claude Code and type /mcp. cloudgpu should show as connected.

Add --scope user to make it available in every project. Replace the placeholders with your own values.

Claude Desktop

  1. Open Settings → Developer → Edit Config. It opens claude_desktop_config.json. Add:
claude_desktop_config.json
{
  "mcpServers": {
    "cloudgpu": {
      "command": "npx",
      "args": [
        "-y",
        "cloudgpu-mcp"
      ],
      "env": {
        "CLOUDGPU_API_KEY": "<YOUR_CLOUDGPU_API_KEY>"
      }
    }
  }
}
  1. Save the file and restart Claude Desktop. Replace the placeholders with your own values.

Needs Node.js on your computer. The file lives in ~/Library/Application Support/Claude/ on macOS and %APPDATA%\Claude\ on Windows.

Cursor

Or add it by hand to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (this project):

mcp.json
{
  "mcpServers": {
    "cloudgpu": {
      "command": "npx",
      "args": [
        "-y",
        "cloudgpu-mcp"
      ],
      "env": {
        "CLOUDGPU_API_KEY": "<YOUR_CLOUDGPU_API_KEY>"
      }
    }
  }
}

Needs Node.js on your computer. Replace the placeholders with your own values.

VS Code

Add it to .vscode/mcp.json. VS Code asks for the secret the first time and stores it securely:

.vscode/mcp.json
{
  "servers": {
    "cloudgpu": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "cloudgpu-mcp"
      ],
      "env": {
        "CLOUDGPU_API_KEY": "${input:cloudgpu-api-key}"
      }
    }
  },
  "inputs": [
    {
      "type": "promptString",
      "id": "cloudgpu-api-key",
      "description": "CLOUDGPU_API_KEY",
      "password": true
    }
  ]
}

Needs Node.js on your computer.

Devin Desktop

  1. Add it to ~/.config/devin/mcp_config.json (macOS and Linux) or %APPDATA%\devin\mcp_config.json (Windows):
mcp_config.json
{
  "mcpServers": {
    "cloudgpu": {
      "command": "npx",
      "args": [
        "-y",
        "cloudgpu-mcp"
      ],
      "env": {
        "CLOUDGPU_API_KEY": "<YOUR_CLOUDGPU_API_KEY>"
      }
    }
  }
}
  1. Refresh the MCP server list in Cascade. Replace the placeholders with your own values.

Devin Desktop is the new name for Windsurf.

Codex

codex mcp add cloudgpu --env "CLOUDGPU_API_KEY=<YOUR_CLOUDGPU_API_KEY>" -- npx -y cloudgpu-mcp

Or edit ~/.codex/config.toml directly:

config.toml
[mcp_servers.cloudgpu]
command = "npx"
args = ["-y", "cloudgpu-mcp"]
env = { CLOUDGPU_API_KEY = "<YOUR_CLOUDGPU_API_KEY>" }

Needs Node.js on your computer. Replace the placeholders with your own values.

Gemini CLI

gemini mcp add -e "CLOUDGPU_API_KEY=<YOUR_CLOUDGPU_API_KEY>" cloudgpu npx -- -y cloudgpu-mcp

This adds it to the current project. Add -s user to use it everywhere.

Any client

Most clients that start local servers accept this shape:

{
  "mcpServers": {
    "cloudgpu": {
      "command": "npx",
      "args": [
        "-y",
        "cloudgpu-mcp"
      ],
      "env": {
        "CLOUDGPU_API_KEY": "<YOUR_CLOUDGPU_API_KEY>"
      }
    }
  }
}

Zed puts servers under context_servers in its settings, with the same command, args and env fields.

Needs Node.js on your computer. Replace the placeholders with your own values.

Embed

Show CloudGPU in a README or on your site. Each embed links to this page, uses the server’s own install command, and sets no cookies. How embeds work

The card, the iframe and the JSON are in English. Add ?lang=it, fr, de or es to their address for another language.

README badge

Badge preview, lightBadge preview, dark
Markdown
[![CloudGPU on mcp.tc](https://mcp.tc/i/cloudgpu/badge.svg)](https://mcp.tc/i/cloudgpu)
HTML
<a href="https://mcp.tc/i/cloudgpu"><img src="https://mcp.tc/i/cloudgpu/badge.svg" alt="CloudGPU on mcp.tc" height="20"></a>

It follows the reader’s light or dark mode. Add ?theme=light or ?theme=dark to fix it, or ?style=compact to show the mark without the word.

Website card

HTML
<script src="https://mcp.tc/w/cloudgpu.js" async></script>

The card appears where the tag is. Add data-theme="dark" or data-size="compact" to the tag to change it. The file holds everything it needs, so it makes no other requests and sets no cookies.

iframe

HTML
<iframe src="https://mcp.tc/embed/cloudgpu" title="CloudGPU on mcp.tc" width="420" height="200" loading="lazy" allow="clipboard-write" style="border:0;border-radius:8px;max-width:100%"></iframe>

Use it on pages that don’t allow scripts. Add ?theme=light or ?theme=dark to the address to fix the colors.

JSON

The listing as data for your own pages and tools: name, tagline, install command, tools and the “Add to” links. Any site can read it (CORS is open).

https://mcp.tc/i/cloudgpu.json

What each field means

AI prompt

Paste this into an AI coding assistant, such as Claude Code or Cursor, and it adds the CloudGPU card to your site.

Markdown
Add the mcp.tc card for CloudGPU, an MCP server, to my website. It is one static script: no build step, no package to install, no cookies.

Put this tag where the card should appear. The card renders right before it:

```html
<script src="https://mcp.tc/w/cloudgpu.js" async></script>
```

To show the card somewhere else, put this placeholder there and load the same script once, for example before `</body>`. Keep the link inside: visitors without JavaScript see it.

```html
<div data-mcptc="cloudgpu"><a href="https://mcp.tc/i/cloudgpu">CloudGPU on mcp.tc</a></div>
```

Options:
- `data-theme="light"` or `"dark"` on the tag or the placeholder fixes the colors. Without it, the card follows the visitor’s light or dark mode.
- `data-size="compact"` gives a smaller card without the tagline.
- The CSS variable `--mcptc-max-width` on the placeholder or a parent sets the width (default `520px`).
- `?lang=` on the script URL sets the card’s language: `en`, `it`, `fr`, `de` or `es`.

If the site has a Content Security Policy, allow `script-src https://mcp.tc` and `img-src data:` (the icon is inline).

For a README or other Markdown docs, use the badge instead:

```markdown
[![CloudGPU on mcp.tc](https://mcp.tc/i/cloudgpu/badge.svg)](https://mcp.tc/i/cloudgpu)
```

Reference: https://mcp.tc/embed

Change history

This server runs on your computer, so we follow its npm package: each new version shows up here, with edits by its owner.

Follow CloudGPU and we’ll email you when something changes.

  1. , Automatic check

    New version 0.16.6 (was 0.16.3). Read from npm.

  2. , Server owner

    Listing updated: description, setup notes and source code.

    Details, 16 Nov 2026, 04:32 UTC

    Description

    Before

    Connects an AI assistant to CloudGPU so it can check live GPU prices, get a quote, rent a GPU by the minute, upload code and data, run commands, tail logs, download results and destroy the instance. It also offers job tools that run on CloudGPU's servers, and lists models on CloudGPU's OpenAI-compatible API. Runs locally over stdio with npx -y cloudgpu-mcp and needs Node.js 18 or newer. Access needs a CloudGPU API key in CLOUDGPU_API_KEY and a prepaid balance. The assistant can also request a key through a browser approval step. Budgets, time limits and idle limits are enforced server-side.

    After

    Connects an AI assistant to CloudGPU so it can check live GPU prices, get a quote, rent a GPU by the minute, upload code and data, run commands, tail logs, download results and destroy the instance. It also offers job tools that run on CloudGPU's servers, and lists models on CloudGPU's OpenAI-compatible API. Runs locally over stdio with npx -y cloudgpu-mcp and needs Node.js 18 or newer. Access needs a CloudGPU API key in CLOUDGPU_API_KEY and a prepaid balance. The assistant can also request a key through a browser approval step. Budgets, time limits and idle limits are enforced server-side. Remote access is also available over Streamable HTTP at https://cloudgpu.app/mcp; the mcp.tc/i/cloudgpu page is a directory listing, not a server endpoint. Local stdio with npx -y cloudgpu-mcp remains supported.

    Setup notes

    Changed. The old and new values aren’t shown.

    Source code

    https://github.com/mrhy7176/cloudgpu-mcp changed to https://github.com/cloudgpu-app/cloudgpu-mcp

  3. , Automatic check

    New version 0.16.3 (was 0.16.1). Read from npm.

  4. , Automatic check

    New version 0.13.1 (was 0.12.1). Read from npm.

  5. , Automatic check

    New version 0.11.0 (was 0.10.0). Read from npm.

  6. , Automatic check

    New version 0.8.0 (was 0.7.0). Read from npm.

  7. , Automatic check

    New version 0.5.0 (was 0.4.0). Read from npm.

  8. , Automatic check

    Tracking started: version 0.4.0.

FAQ

Can I paste mcp.tc/i/cloudgpu into my MCP client?

No. CloudGPU runs on your own computer, started by your client, so it has no web address to connect to. The quick link is the page to share; the install command is npx -y cloudgpu-mcp.

Does CloudGPU need an API key?

Yes. It reads CLOUDGPU_API_KEY from its environment, and CLOUDGPU_API_KEY is a secret. Put the value in your client’s config on your own machine, never in a shared file.

Is CloudGPU a remote or a local server?

Local. Your client starts it as a process on your computer with npx -y cloudgpu-mcp, which needs Node.js.

What can CloudGPU do?

You can list GPU models with live prices and availability, get a cost quote before renting and create and stop GPU instances with budget and time limits.

Which clients can use it?

Any client that starts local servers: Claude Code, Claude Desktop, Cursor, VS Code, Devin Desktop, Codex, Gemini CLI, Zed and others. claude.ai and ChatGPT only connect to remote servers.

Does mcp.tc track changes to CloudGPU?

Partly. It runs on your computer, so we follow its npm package: its History tab shows each new version and edits by its owner. Follow the listing to get an email when it changes.

Who wrote this page?

mcp.tc’s robot read CloudGPU’s public metadata (its package and repository pages), and an AI model drafted the text from it. A person reviews anything the checks can’t confirm. The text can still be wrong, so if you spot a mistake, use Report this listing on this page.

mcp.tc isn’t affiliated with mrhy7176. This page was built from CloudGPU’s public metadata, last checked on 8 Nov 2026, and an AI model wrote the description, so it can be wrong. Names and marks belong to their owners. Something off? Report this listing.