Skip to content

MCP

Connect once. Then ask in plain words.

MCP lets Claude Code, Codex or Cursor use outside tools. Add Runcrate with one command and your agent launches GPUs, runs your job on them and stops the bill, from $0.20/hr.

You sign in through your browser. Your agent never holds a key.

Your agent launches it over MCP21 tools

Connect

One command, then sign in. No key to copy first.

  1. 01

    Add Runcrate.

    One command in Claude Code or Codex, or one entry in Cursor. Nothing to install.

  2. 02

    Sign in when it asks.

    Your browser opens the Runcrate sign-in. Approve, and your agent works in your workspace.

  3. 03

    Ask.

    In plain words, the way you would ask a person: “Launch an H100 with the Finetune crate and run nvidia-smi.”

terminalbash
claude mcp add --transport http runcrate https://mcp.runcrate.ai/mcp

Then run /mcp in Claude Code and pick runcrate: your browser opens to sign in. No key to paste.

Any other MCP client that signs in with OAuth needs only the address: https://mcp.runcrate.ai/mcp

Tools

21 tools your agent can use. It picks the right one for each request.

  • Instances

    See prices, launch a GPU on a crate, check on it, and stop it.

    • list_gpu_types
    • create_instance
    • list_instances
    • get_instance
    • instance_status
    • delete_instance
  • Run and files

    Run a command on a running instance, and copy small files to and from it, such as a script or a results file.

    • ssh_execute
    • file_upload
    • file_download
  • Storage

    Volumes that keep your files when an instance stops.

    • list_volumes
    • create_volume
    • delete_volume
    • list_storage_regions
  • Billing

    Your balance, and what you spent on what.

    • billing_balance
    • billing_usage
  • Crates and SSH keys

    The ready images an instance can boot, and the keys for your own ssh.

    • list_templates
    • list_ssh_keys
    • add_ssh_key
  • Environments

    Keep staging apart from production inside one workspace.

    • list_environments
    • create_environment
    • delete_environment
Every tool and its parameters

Examples

Ask the way you would ask a person. Real prompts from the docs, and the tools they call.

  1. “Deploy an A100, upload my dataset, and start training.”

    • list_gpu_types
    • create_instance
    • instance_status
    • file_upload
    • ssh_execute
    The walkthrough
  2. “Compare A100, H100, and RTX 4090 pricing. I need at least 40GB VRAM.”

    • list_gpu_types
    The walkthrough
  3. “Set up a vLLM server with Llama 3.1 70B on an A100.”

    • create_instance
    • instance_status
    • ssh_execute
    • get_instance
    The walkthrough
  4. “My training script is crashing with CUDA OOM. Can you check what's happening?”

    • list_instances
    • get_instance
    • ssh_execute
    The walkthrough
  5. “Find anything that's running but not being used and shut it down.”

    • list_instances
    • ssh_execute
    • delete_instance
    The walkthrough

Sign-in and safety

You sign in. Your agent never holds a key.

  • A browser sign-in, never a pasted key

    Your client registers itself and opens the Runcrate sign-in page. Your password never passes through the agent, and no key goes in a config file.

  • Every tool says what it does

    Each tool is marked read-only, creating, or deleting, so your client can ask you before your agent stops an instance or deletes a volume.

  • Your workspaces, your environments

    Your agent acts only on workspaces you belong to, and every tool can be pointed at one environment, such as staging.

  • 100 requests a minute

    Far more than a person asking needs, and a stop on a runaway loop.

Pricing

The prices in your dashboard. Your agent pays the same rate you do.

Billed by the second on most GPUs while an instance runs; a few bill by the minute. Delete it and the bill stops.

See every GPU price
H100$3.25/hrFrom, live
Cheapest GPU-hour$0.20/hrA4000, live
GPU models19On offer today

FAQ

Questions about MCP.