LeeQ

Experiment orchestration and autonomous calibration for superconducting quantum systems.

LeeQ is a Python framework for defining, executing, and analyzing pulse-level quantum experiments. It was built for superconducting-circuit research, where experiments must connect reusable pulse sequences, parameter sweeps, calibration state, hardware control, simulation, and scientific analysis.

Unlike a circuit-level SDK, LeeQ focuses on the laboratory workflow around the quantum processor: characterize a device, update its calibration, preserve the experimental context, and compose the next experiment.

Why LeeQ?

• Experiment-native API: express pulse sequences, measurements, and multidimensional parameter sweeps as reusable Python objects.
• Calibration workflows: run spectroscopy, Rabi, Ramsey, DRAG, readout calibration, benchmarking, tomography, and multi-qubit tune-up routines.
• Simulation-to-hardware path: develop against virtual transmon models, then connect the same experiment abstractions to laboratory control systems, including QubiC.
• Agent-ready experiments: expose structured experiment descriptions and text/visual inspection hooks through the k-agents integration.
• Reproducible sessions: track experiment inputs, outputs, and calibration changes with the integrated Chronicle layer.

Quick start

Run the Docker environment

docker run --rm \
  -p 8888:8888 \
  -p 8050:8050 \
  -v /path/to/local/folder:/home/jovyan/work \
  ghcr.io/shuxiangcao/leeq:latest

Open http://localhost:8888 for Jupyter. Port 8050 is available for live plotting.

Install from source

LeeQ requires Python 3.10 or newer.

git clone https://github.com/ShuxiangCao/LeeQ.git
cd LeeQ
python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e .

Then follow the 10-minute simulated experiment to run a Rabi calibration without laboratory hardware.

How it fits together

LeeQ keeps experiment logic separate from execution backends. This makes it possible to develop and test workflows in simulation while preserving the abstractions needed for real instruments.

Included workflows

Area	Examples
Single-qubit calibration	resonator and qubit spectroscopy, Rabi, Ramsey, DRAG, ping-pong amplitude refinement
Characterization	T1, T2, randomized benchmarking, assignment matrices
Multi-qubit control	conditional Stark calibration, Hamiltonian tomography, two-qubit tune-up
Analysis	fitting, plotting, tomography, optimal control, inspection hooks
Infrastructure	virtual devices, QubiC compilation, and Chronicle session tracking

Explore the runnable notebooks:

• Simulation and tune-up
• Calibration agents
• Tutorial series
• Characterization and calibration workflows
• Focused experiment examples

Documentation

• Documentation site
• Installation guide
• Core concepts
• Experiment guide
• Calibration guide
• Architecture overview

Development

python -m pip install -r requirements-dev.txt
python -m pip install -e .
pytest tests/

See the contributing guide for the development workflow. Pull requests and issue reports are welcome.

Project status

LeeQ is research software under active development. It has been used to develop and operate superconducting-qubit experiments, but APIs may continue to evolve as new hardware and autonomous-experimentation workflows are added.

Acknowledgement

LeeQ was created by Shuxiang Cao at the University of Oxford's Quantum Superconducting Circuits Research Group.

License

LeeQ is available under the BSD 3-Clause License.