VRAMfarm GPU Cloud
Rent EU cloud GPUs by the minute from an agent, with a hard hourly budget per launch
Answered our latest tool-list check 15 hours ago with 7 tools. It will be marked verified after 3 successful checks in 7 days.
- Pricing
- Paid
- Platforms
- Web, API, CLI
- Runs
- Remotely
- Version
- 1.0.1
- Listing updated
- 3 days ago
About VRAMfarm GPU Cloud
VRAMfarm GPU Cloud is a paid MCP server for cloud and infrastructure. It runs as a hosted remote server, so there is nothing to install locally.
VRAMfarm rents GPU machines billed by the minute from a prepaid euro balance, and this server lets an agent run the whole cycle: compare cards by VRAM and price, pick a concrete machine, launch it, read the SSH command and app links, and stop it. Launch templates include PyTorch, ComfyUI, chat, vLLM and text to speech, or you can supply your own Docker image.
Spending is capped in several ways. Each launch must carry max_eur_per_hour, and a machine above that budget is refused before anything is rented. The balance is prepaid, so an agent cannot spend more than you topped up, billing has a 15-minute minimum, and idle servers get an email and auto-stop. The API token can only list machines and launch, check and stop servers, not change passwords, keys or payments.
Prices on the pricing page start at 0.12 euros an hour for an RTX 3060 and reach 10.90 euros for a B200, with internet traffic billed separately.
Install
Run in a terminal
claude mcp add --transport http gpu-cloud https://vramfarm.com/mcpAdd to Settings → Connectors → Add custom connector
https://vramfarm.com/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"gpu-cloud": {
"url": "https://vramfarm.com/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"gpu-cloud": {
"type": "http",
"url": "https://vramfarm.com/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"gpu-cloud": {
"serverUrl": "https://vramfarm.com/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"gpu-cloud": {
"type": "http",
"url": "https://vramfarm.com/mcp"
}
}
}Tools this server provides (7)
Read directly from the server.
- list_gpu_types
- One row per GPU model and card count: the cheapest reliable machine, its price in EUR per hour, VRAM per card and how many are available. Start here to choose a GPU.
- list_offers
- Concrete machines of one GPU model, cheapest first (up to 10), with country, CUDA, disk, network and price. Use the id as offer_id in launch_server to pick an exact machine.
- launch_server
- Rents a GPU machine and starts a template or a Docker image on it. Costs money from the prepaid balance, billed per minute while running (minimum 15 minutes). Returns at once with the server id; the server is ready in 2–5 minutes (8–15 with a custom image). Refuses machines priced above max_eur_per_hour.
- list_servers
- Your servers, newest first: running, starting and stopped in the last 7 days, with status, price and access.
- get_server
- One server: status (provisioning, running, stopped), what it is doing now, cost so far, and once running the access: SSH command, Jupyter link with token, app URL.
- stop_server
- Stops a server and its billing. The disk is wiped; with save_workspace true the /workspace folder is saved first (storage is billed per GB-month) and comes back on the next launch.
- get_account
- Balance in EUR, how long it lasts at the current spend, whether an SSH key is set, and limits.
What it does
- GPU catalogue
- Lists models with VRAM, cheapest hourly price and free machine count.
- Budget cap per launch
- Refuses machines above the agent's max hourly price.
- Ready templates
- Starts PyTorch, ComfyUI, chat, vLLM or text to speech, or your own Docker image.
- SSH and app links
- Returns the SSH command, Jupyter and app links for a running server.
- Workspace saving
- Can save /workspace when stopping for the next launch.
Good for
- Find the cheapest 24 GB GPU under a set price
- Launch ComfyUI on an RTX 4090 and stop it later
- Run a training script on an A100 and stop when done
- Check balance and remaining runtime
Pricing
Paid
GPU rental from EUR 0.12/hour, billed per minute from prepaid balance (top up from EUR 10); traffic billed separately
Prices change. Check vramfarm.com before you buy.
VRAMfarm GPU Cloud FAQ
How is it billed?
Per minute from a prepaid balance after a 15-minute minimum; top-ups start at EUR 10.
Does MCP cost extra?
No. You pay only for servers the agent runs at dashboard prices.
What can the token do?
List machines, and launch, check and stop servers. It cannot change passwords, keys or payments.
Can I use REST instead?
Yes. The MCP tools are a thin layer over the same REST API and token.
Is VRAMfarm GPU Cloud free?
No. VRAMfarm GPU Cloud is a paid product.
How do I add VRAMfarm GPU Cloud to Claude Code?
Run this command in your terminal: claude mcp add --transport http gpu-cloud https://vramfarm.com/mcp
What are the best alternatives to VRAMfarm GPU Cloud?
The closest alternatives to VRAMfarm GPU Cloud are Claude Code Ultimate Guide, Agent Yurt, Proofline, signafAIde and Toolkien.
Alternatives to VRAMfarm GPU Cloud

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

Agent Yurt
MCP serverBack up agent memory, files and dashboards, then recover from a reviewed snapshot
Not checked yet- Remote
- Free
- Knowledge and memory

Proofline
MCP serverProduction-aware review of changes against your AWS or Google Cloud setup, via MCP
- Remote
- Freemium
- Developer tools

signafAIde
MCP serverLocal MCP server to set up, verify and check coverage of signed agent action records
Runs locally- 5 tools
- Freemium
- Security

Toolkien
MCP serverOne MCP gateway for all your MCP servers, with code mode, permissions and audit
Not checked yet- 3 tools
- Remote
- Open source
- AI models and LLM tooling

filex
MCP serverSelf-hosted file manager in one Go binary with a built-in MCP server for agents
Not checked yet- Remote
- Open source
- Files and documents