Tutorials

Learn how quantum compilation works by predicting a result, inspecting the compiler’s output, and changing the program. These six executable notebooks assume basic Python and quantum computing knowledge; no MLIR experience is needed.

  1. Follow a program through the compiler: follow QC and QCO, remove redundant gates, and compare unitary matrices.

  2. Follow registers and control flow: trace registers, loops, and measurement-driven branches, then check their sampled results.

  3. Compile for hardware constraints: compare hardware constraints, inspect routing, and submit a compiled program through QDMI.

  4. Execute programs with QIR: inspect Base and Adaptive QIR, execute feedback, and relate output records to counts.

  5. Run programs through QDMI: discover device capabilities, submit and reuse jobs, and compare shots, counts, and simulator state results.

  6. Exchange structured programs with jeff: hand a structured program between compiler processes, preserve phase and loops, then compile it for execution.

For a first execution, use the QPE walkthrough. The compiler guide describes interfaces and options; these tutorials explain their effects through experiments.

Run the notebooks

Install uv. Download requirements.txt and use the download link at the start of each tutorial to save its notebook in the same directory.

In that directory, start JupyterLab with all tutorial dependencies:

uv run --no-project --with-requirements requirements.txt jupyter lab

uv manages and caches the environment. This command installs MQT Core, JupyterLab, and the supported Qiskit visualization dependencies together. Open a notebook with the Python 3 kernel and run its cells from top to bottom. Each notebook has its own imports and parameters. See uv’s Jupyter guide to use an existing project or an editor instead.

MQT Core’s Python wheels include the compiler and local simulator; no separate MLIR installation or hardware account is needed. See Installation for platform requirements and source builds, and Qiskit compatibility for supported Qiskit versions.

On this website, cells and figures show results from the documentation build. Expand Show code cell source to inspect setup and plotting cells. These cells remain in the downloaded notebooks; edit the parameter cells and rerun to try the experiments.