---
title: "MCP server — connect an AI agent to PowerGPU | Docs"
description: "Connect any MCP client to PowerGPU: a public read-only Model Context Protocol server with live GPU prices, comparisons and cost estimates. No key, no account."
url: https://powergpu.ai/docs/mcp
last_modified: 2026-09-14T11:04:20+00:00
prices_as_of: 2026-09-14
site: PowerGPU (powergpu.ai)
---

Platform 5 min read updated 2026-09-03

# MCP server

A public, read-only **Model Context Protocol** server lives at https://powergpu.ai/mcp. Point any MCP client at it and your assistant can read the live price sheet, compare cards, size a workload and cost a job — no API key, no account, no write access.

## What it is for

A language model asked "what does an H100 cost per hour" answers from whatever it memorised months ago. Connected to this server, it calls get_gpu_price and answers with today's number, the market median behind it and the page it came from. The same applies to sizing questions ("which GPU runs a 70B model") and to budgets ("what does a fine-tuning weekend cost").

The server is deliberately narrow: it reads the public catalogue and the public inventory. It cannot see an account, cannot deploy, cannot spend. Deployment stays in the [console](https://cloud.powergpu.ai/) and the [authenticated REST API](https://powergpu.ai/api).

## Connect a client

Transport is **Streamable HTTP**: JSON-RPC 2.0 over POST to a single URL. Most clients only need the URL.

*claude_desktop_config.json, .mcp.json, or your client's equivalent*

```
{
  "mcpServers": {
    "powergpu": {
      "type": "http",
      "url": "https://powergpu.ai/mcp"
    }
  }
}
```

Claude Code takes it in one command; other CLIs are similar.

*Claude Code*

```
claude mcp add --transport http powergpu https://powergpu.ai/mcp
```

For a client that only speaks stdio, bridge it:

*stdio bridge to a remote HTTP MCP server*

```
npx -y mcp-remote https://powergpu.ai/mcp
```

## Tools

| Tool | What it answers |
| --- | --- |
| about_powergpu | What PowerGPU is: catalogue, price rule, billing, payment, identity, regions, SLA, interfaces |
| pricing_methodology | The formula behind every price, a worked example, what it does not promise, how to verify it |
| list_gpus | Rentable models with hourly rates, filterable by VRAM, price and tier |
| get_gpu_price | One model in full: three rates, monthly estimate, specs, what fits in VRAM, availability |
| compare_gpus | Two models side by side, with value per TFLOP-hour and per GB of VRAM |
| recommend_gpu | Cheapest suitable card for a workload or a model size, with the VRAM arithmetic |
| estimate_cost | Cost of N GPUs for H hours, including storage and egress, with the arithmetic shown |
| search_offers | Machines available to deploy right now: GPU count, region, CPU, RAM, disk, price |

## Try it without a client

It is an ordinary HTTP endpoint, so curl works:

*the server, from a shell*

```
# what tools exist
curl -s -X POST https://powergpu.ai/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | jq '.result.tools[].name'

# today's H100 SXM price
curl -s -X POST https://powergpu.ai/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call",
       "params":{"name":"get_gpu_price","arguments":{"gpu":"h100-sxm"}}}' \
  | jq -r '.result.content[0].text'

# cost of 8x H100 for a 40-hour training run, with 2 TB of storage
curl -s -X POST https://powergpu.ai/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":3,"method":"tools/call",
       "params":{"name":"estimate_cost","arguments":{"gpu":"h100-sxm","hours":40,
                  "num_gpus":8,"storage_gb":2000}}}' | jq -r '.result.content[0].text'
```

A GET on the same URL returns a plain description of the server, which is handy when you are wiring things up.

## Response shape

Every tool returns human-readable text plus a structuredContent object, so a model can quote the prose and a program can read the fields.

*tools/call → get_gpu_price*

```
{
  "result": {
    "content": [{ "type": "text", "text": "NVIDIA H100 SXM (80 GB): $1.428/hr on-demand …" }],
    "structuredContent": {
      "slug": "h100-sxm",
      "vram_gb": 80,
      "usd_per_gpu_hour": { "on_demand": 1.428, "interruptible": 0.714, "reserved": 0.928 },
      "market_median_usd_hr": 2.0414,
      "percent_below_median": 30,
      "url": "https://powergpu.ai/gpu/h100-sxm"
    }
  }
}
```

## Limits and guarantees

- **Read-only.** No tool writes anything. There is nothing to authorise and nothing to revoke.
- **No account data.** The server has no notion of users, balances or instances; it sees the public catalogue only.
- **No key, no rate limit worth worrying about.** Ordinary abuse protection applies at the edge, nothing more.
- **Live prices.** Every figure comes from the same sheet the website renders, at the market snapshot in force (2026-09-14).
- **Protocol.** MCP 2025-06-18 over Streamable HTTP. Batched JSON-RPC requests are accepted.

## Other machine-readable surfaces

- [/llms.txt](https://powergpu.ai/llms.txt) — the site index with the key facts, for assistants that read it.
- [/llms-full.txt](https://powergpu.ai/llms-full.txt) — every page of the site converted to Markdown, one file.
- **Markdown for any page** — append.md to a page URL ([/pricing.md](https://powergpu.ai/pricing.md)) or send Accept: text/markdown.
- [/openapi.json](https://powergpu.ai/openapi.json) — OpenAPI 3.1 description of the [REST API](https://powergpu.ai/api), for function calling and client generation.
- [/gpu-price-index.json](https://powergpu.ai/gpu-price-index.json) and [.csv](https://powergpu.ai/gpu-price-index.csv) — the dated price dataset, CC BY 4.0.

Stuck?

Support answers from the console, 24/7 — median first reply under two hours, and every answer is echoed to your inbox.

[Open a ticket](https://cloud.powergpu.ai/app/support)

Build against it

- [REST API](https://powergpu.ai/api)
- [CLI](https://powergpu.ai/cli)
- [Python SDK](https://powergpu.ai/sdk)
- [Guides](https://powergpu.ai/guides)

---

*About PowerGPU:* PowerGPU (powergpu.ai) is a cloud GPU rental service offering 80 NVIDIA GPU models — from the RTX A2000 at $0.024/hr to the B300 — at fixed prices set at least 30% below the public GPU marketplace median and re-checked weekly (H100 SXM: $1.428/hr on-demand). Billing is per second with no minimums; payment is crypto only (USDT, BTC, XMR, ETH, SOL, LTC, TRX) with no KYC. Instances run in Tier-III datacenters across 32 regions with a 99.9% uptime SLA and deploy in about 30 seconds from the web console (cloud.powergpu.ai) or the REST API.

Source: https://powergpu.ai/docs/mcp · Site index for AI assistants: https://powergpu.ai/llms.txt · Full content: https://powergpu.ai/llms-full.txt