Quantum Hardware MCP Server
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Quantum Hardware MCP ServerWhich quantum computer has the shortest queue?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Quantum Hardware MCP Server
An MCP server that gives AI assistants (Claude, etc.) live data about IBM Quantum computers — queue depths, error rates, coherence times, and more.
Built with the MCP Python SDK and qiskit-ibm-runtime.
Tools exposed
Tool | What it does |
| All IBM quantum computers you can access + status |
| Deep info on one machine: error rates, T1/T2, queue |
| Rank machines by CX error, queue depth, or qubit count |
| Current queue snapshot — useful for picking the shortest wait |
Related MCP server: qiskit-sim-mcp
Prerequisites
Python 3.10 or newer
An IBM Quantum account (free) — sign up at quantum.ibm.com
Claude Desktop (to use the MCP integration)
Setup
1. Clone the repo
git clone https://github.com/YOUR_USERNAME/quantum-hardware-mcp.git
cd quantum-hardware-mcp2. Create a virtual environment
python3 -m venv venv
source venv/bin/activate # macOS / Linux
# venv\Scripts\activate # Windows3. Install dependencies
pip install -r requirements.txt4. Get your IBM Quantum API token
Go to quantum.ibm.com/account
Log in (or create a free account)
Copy your API token from the "API token" section
5. Create your .env file
cp .env.example .envOpen .env and replace your_token_here with your real token:
IBM_QUANTUM_TOKEN=abc123...your_actual_token...xyzImportant:
.envis in.gitignore. It will never be committed. Never paste your token directly intoserver.py.
6. Test the server runs
python server.pyYou should see no errors. Press Ctrl+C to stop it.
The server speaks over stdin/stdout (MCP stdio transport), so it won't print anything — a clean exit means it's working.
Connect to Claude Desktop
1. Find your Claude Desktop config file
OS | Path |
macOS |
|
Windows |
|
2. Add the server entry
Open the file (create it if it doesn't exist) and add:
{
"mcpServers": {
"quantum-hardware": {
"command": "/absolute/path/to/venv/bin/python",
"args": ["/absolute/path/to/quantum-hardware-mcp/server.py"]
}
}
}Replace the paths with your actual paths. To find them:
# From inside the project folder with venv activated:
which python # → path to use for "command"
pwd # → prefix for "args" pathExample on macOS:
{
"mcpServers": {
"quantum-hardware": {
"command": "/Users/yourname/quantum-hardware-mcp/venv/bin/python",
"args": ["/Users/yourname/quantum-hardware-mcp/server.py"]
}
}
}3. Restart Claude Desktop
Quit and reopen Claude Desktop. You should see "quantum-hardware" in the MCP tools list (hammer icon).
Usage examples
Once connected, ask Claude:
"List all IBM quantum computers I can access"
"Get details for ibm_brisbane"
"Which quantum computer has the lowest error rate right now?"
"Which machine has the shortest queue right now?"
"Compare all devices by queue depth"
Project structure
quantum-hardware-mcp/
├── server.py # MCP server — all tool logic lives here
├── requirements.txt # Python dependencies
├── .env.example # Token template (safe to commit)
├── .env # Your real token (git-ignored, never commit)
├── .gitignore
├── LICENSE # MIT
└── README.mdHow it works (quick mental model)
Claude Desktop
│ asks a question
▼
MCP Protocol (stdin/stdout)
│ calls a tool
▼
server.py ──── QiskitRuntimeService ────► IBM Quantum API
│ │
│ ◄─── live device data ─────────────────┘
▼
Claude Desktop (formats and answers)The MCP protocol is like a plugin system: Claude Desktop spawns server.py as a child process and sends JSON messages over stdin/stdout. The server responds with tool results. No HTTP server needed.
Troubleshooting
Problem | Fix |
| Check your |
| Token is wrong or expired — get a fresh one from quantum.ibm.com |
Server not showing in Claude Desktop | Check the paths in |
| Normal — it makes one API call per device to get calibration data |
License
MIT — see LICENSE.
This server cannot be installed
Maintenance
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Lokesh-2025/quantum-hardware-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server