qrmfold

qrmfold (Quantum Reed-Muller Fold) is the Python package accompanying the paper Construction of the full logical Clifford group for high-rate quantum Reed-Muller codes using only transversal and fold-transversal gates. Its main function is to output addressable S, Hadamard, controlled-Z, and swap gates for quantum Reed-Muller codes. The output format is a stim.Circuit which can then easily be converted to QASM.

Installation Instructions

qrmfold is available as a PyPI package so it can be installed by running pip install qrmfold.

Tutorial

See the getting_started.ipynb notebook. For more detail on individual functions/methods, see below.

Documentation

If you just want to read the documentation, click the blue Documentation badge at the top of this README.

Optionally, you can build the HTML docs locally:

• Install docs dependencies: pip install -e ".[docs]"
• Build: cd docs && make html
• Open: docs/build/html/index.html

How to cite qrmfold

Please cite the accompanying paper:

@misc{tansuwannont2026constructionlogicalcliffordgroup,
      title={Construction of the full logical Clifford group for high-rate quantum Reed-Muller codes using only transversal and fold-transversal gates}, 
      author={Theerapat Tansuwannont and Tim Chan and Ryuji Takagi},
      year={2026},
      eprint={2602.09788},
      archivePrefix={arXiv},
      primaryClass={quant-ph},
      url={https://arxiv.org/abs/2602.09788}, 
}

Tests

This package uses the pytest framework. The tests are divided into unit tests and integration tests. The integration tests verify that:

1. the fold-transversal gates preserve the stabilizer group (Theorems 3 and 4),
2. the fold-transversal gates induce the logical action predicted by Theorems 5 and 6,
3. the addressable gates built from Corollary 1 and Theorems 7 and 8 induce the desired logical action.

Note there is one test, test_quantum_reed_muller_integration.TestAddressableLogicalAction.test_2_qubit_gate for the parameter $m =6$, that takes considerably longer than the others i.e. about 300 s on a laptop.

Release (maintainers)

This repo includes a GitHub Actions workflow that publishes to PyPI when a version tag like v0.7.0 is pushed.