# ExducerAI GPU cloud

> ExducerAI GPU cloud is a paid MCP server for cloud and infrastructure. Rent whole RTX PRO 6000 GPUs in Montréal from Claude Code, Codex or Cursor. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, CLI
- **Website:** https://exducer.ai/agents
- **MCP Registry name:** ai.exducer/turbo
- **Categories:** [cloud and infrastructure](https://ai-powered.com/categories/cloud-and-infrastructure)
- **Listing updated:** 2026-11-19

## About

The ExducerAI MCP server lets an AI coding agent handle the full lifecycle of a GPU rental: checking availability and list prices, quoting a booking with tax included, booking, waiting for the machine to start, connecting over SSH, and stopping the deployment. The server is one of three pieces installed by a setup prompt, alongside a command-line client and agent skills that teach the workflow.

The README and site describe a confirmation gate: the agent can read prices, quotes and deployment costs on its own, but booking, extending or enabling automatic renewal requires an explicit yes from the user, and stopping a deployment deletes its container and disk. API, SSH and storage keys are created only in the console.

A sandbox key runs the same flow against a simulated lifecycle with no charges, which suits developers who want to test an agent workflow before renting real hardware. The service is aimed at teams who want GPU capacity hosted in Canada, with list prices in CAD and no egress fees.

## Features

- **Quote before booking:** Agents fetch a tax-inclusive quote and must get user confirmation before any booking is made.
- **Spending confirmation gate:** The server refuses booking calls that do not state the user has confirmed the purchase.
- **Sandbox key for testing:** A sandbox key runs the full flow against a simulated lifecycle with no charges.
- **Idempotent bookings:** Bookings and extensions carry idempotency keys so a retry does not book twice.
- **SSH host key verification:** The client refuses to connect to a machine whose host key differs from the one published by the API.
- **Keys kept out of the agent:** The API key stays in the CLI configuration and is not copied into the agent's settings.

## Good for

- Renting a whole RTX PRO 6000 GPU for a training or inference run started from a coding agent
- Comparing GPU list prices and availability before committing to a booking
- Testing an agent-driven GPU workflow with a sandbox key before spending money
- Stopping a deployment and copying results off the disk before it is deleted

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http turbo https://exducer.ai/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://exducer.ai/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "turbo": {
      "url": "https://exducer.ai/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "turbo": {
      "type": "http",
      "url": "https://exducer.ai/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "turbo": {
      "serverUrl": "https://exducer.ai/mcp"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://exducer.ai/mcp
```

Enter a Name (for example "ExducerAI GPU cloud"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "turbo": {
      "type": "http",
      "url": "https://exducer.ai/mcp"
    }
  }
}
```

## Tools

18 of 22 tools say they're read-only. 4 of 22 tools say they can delete or change data. From the server's own tool annotations; not verified by us.

- `get_pricing` (says read-only): List prices in CAD: GPU-hour per GPU model, object storage per GB-month, and the Models API. Use this instead of any remembered price.
- `check_availability` (says read-only): One yes or no per GPU model: is it free right now. A no does not block a request; the request waits for capacity.
- `get_service_status` (says read-only): The public status of each service (console, API, GPU region, object storage, Models API).
- `list_servers` (says read-only): The servers a booking can name: id (the server_id of quote_booking and create_booking), GPU model, price per GPU-hour in CAD, allowed_gpu_counts, region, and whether it is free now.
- `list_templates` (says read-only): The templates a booking can name: slug, name, the container image it runs, category. Without a template a deployment runs the default PyTorch image.
- `quote_booking` (says read-only): What a booking would cost in CAD: GPU cost, tax and total for the whole duration, and the hourly rates. Books nothing and charges nothing. Always quote and show the user the total before create_booking.
- `search_docs` (says read-only): Search the documentation by keyword. Returns the matching sections with a snippet and the slug to pass to read_doc.
- `read_doc` (says read-only): One documentation page as Markdown, by its slug (from search_docs). Read the page before answering a question about how the platform works.
- `get_account` (says read-only): The account behind the API key (email, company, type), whether the key is a sandbox key, and the prepaid balance in CAD (simulated under a sandbox key).
- `list_deployments` (says read-only): Every deployment of the account with its state, GPU count, rate, cost so far and end time. States: requested, queued, provisioning, running, then stopped, ended, failed or cancelled.
- `get_deployment` (says read-only): One deployment in full: state and reason, SSH host and port, the host key fingerprint to verify, rate per GPU-hour, cost so far, start and end time, endpoints.
- `get_deployment_events` (says read-only): The state changes of a deployment, oldest first, each with its reason, actor and time. Read this to learn why a deployment failed, ended or is waiting.
- `wait_for_deployment` (says read-only): Poll a deployment until it reaches a state (default running) or a final state, for at most 50 seconds per call. Returns reached, the current state and whether it is final. While reached is false and final is false, call again; a live request can stay requested until staff place it, so after a few minutes tell the user and check back later instead of looping.
- `get_billing` (says read-only): The prepaid balance and the most recent credit ledger lines (top-ups and charges) in CAD.
- `get_usage` (says read-only): What the account was charged for GPU time, object storage and the Models API over a date range, in CAD before tax. Only deployments that ran appear.
- `list_invoices` (says read-only): The account's invoices (not available to a sandbox key).
- `list_ssh_keys` (says read-only): The SSH public keys saved on the account: a booking with no ssh_public_key uses them (not available to a sandbox key). Adding or removing keys is done by the user in the console.
- `get_storage_usage` (says read-only): Object storage: GB stored, active keys, price per GB-month and the amount billed this month (not available to a sandbox key).
- `create_booking` (says it can change data): Send a GPU request. It is created as requested or queued and becomes running later (use wait_for_deployment); nothing is charged before it runs, then credit is drawn per minute. Quote first with quote_booking and get the user's yes on the total. Retry-safe: see idempotency_key.
- `extend_deployment` (says it can change data): Add hours to the term of a deployment that is requested, queued, provisioning or running. Same machine, same SSH details; billing stays per minute. Tell the user the extra cost (hours × gpu_count × rate) and get a yes first. Retry-safe: see idempotency_key.
- `set_auto_renew` (says it can change data): Turn automatic renewal of a deployment on or off. On, it keeps running and drawing credit past its end time until it is turned off or credit runs short, so get the user's yes first.
- `stop_deployment` (says it can change data): Stop a deployment now: billing ends at this minute and its container and NVMe scratch disk are deleted and cannot be recovered. A request that has not started is withdrawn at no cost. Make sure results are copied off, tell the user the disk is deleted, and get a yes first.

## FAQ

### Which coding agents does the setup support?

The setup command configures Claude Code, Codex and Cursor. Other MCP clients can connect to the hosted server with an API key as a Bearer header or run the server over stdio through the CLI.

### Does the agent need my API key?

The key stays in the CLI's own configuration, is not copied into the agent's settings, and the agent is told never to ask for it in a chat.

### Is there a free way to try it?

A sandbox key runs the full flow against a simulated lifecycle. Nothing is charged and no machine is involved.

### Does stopping a deployment keep my data?

Stopping a deployment deletes its container and its disk, so the documentation advises copying results off before a stop.

### Is ExducerAI GPU cloud free?

No. ExducerAI GPU cloud is a paid product.

### How do I add ExducerAI GPU cloud to Claude Code?

Run this command in your terminal: claude mcp add --transport http turbo https://exducer.ai/mcp

### What are the best alternatives to ExducerAI GPU cloud?

The closest alternatives to ExducerAI GPU cloud are GPU Cloud, CloudGPU MCP, Xcelsior GPU Cloud, Center AI Vision GPU cloud and CloudGPU MCP.

## Where this comes from

- Description: written from the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [GPU Cloud](https://ai-powered.com/mcp/gpu-cloud): GPU Cloud is a paid MCP server for cloud and infrastructure.
- [CloudGPU MCP](https://ai-powered.com/mcp/cloudgpu-mcp): CloudGPU MCP is a paid MCP server for cloud and infrastructure.
- [Xcelsior GPU Cloud](https://ai-powered.com/mcp/xcelsior-gpu-cloud): Xcelsior GPU Cloud is a paid MCP server for cloud and infrastructure.
- [Center AI Vision GPU cloud](https://ai-powered.com/mcp/center-ai-vision-gpu-cloud): Center AI Vision GPU cloud is a paid MCP server for cloud and infrastructure.
- [CloudGPU MCP](https://ai-powered.com/mcp/cloudgpu-mcp-2): CloudGPU MCP is a paid MCP server for cloud and infrastructure.
- [ParalonCloud](https://ai-powered.com/mcp/paraloncloud): ParalonCloud is a paid MCP server for cloud and infrastructure.
- [GPUvec](https://ai-powered.com/mcp/gpuvec): GPUvec is a free MCP server for cloud and infrastructure.
- [gpu.nz GPU cloud](https://ai-powered.com/mcp/gpu-nz-gpu-cloud): gpu.nz GPU cloud is a paid MCP server for cloud and infrastructure.

---
Source: https://ai-powered.com/mcp/exducerai-gpu-cloud
