RydOpt - A Multiqubit Rydberg Gate Optimizer

RydOpt is a Python package for the optimization of laser pulses implementing two- and multiqubit Rydberg gates in neutral atom quantum computing platforms. The optimization methods support GPUs and multi-core CPUs, using an efficient implementation based on JAX.

Install the software with pip (requires Python ≥ 3.10; for enabling GPU support and tips, see our extended installation instructions):

pip install rydopt

Documentation and Minimal Usage Example

The package documentation can be found at https://rydopt.readthedocs.io/.

To get an idea how the software is used, we provide in the following a minimal usage example. The code optimizes a pulse to realize a CZ gate on two atoms in the perfect blockade regime.

import rydopt as ro
import numpy as np

# Want to perform a CZ gate on two atoms in the perfect blockade regime; no Rydberg state decay
gate = ro.gates.TwoQubitGate(phi=None, theta=np.pi, Vnn=float("inf"), decay=0.0)

# Pulse ansatz: constant detuning, sweep of the laser phase according to a sine CRAB ansatz
pulse_ansatz = ro.pulses.SinglePhotonPulseAnsatz(detuning_ansatz=ro.pulses.Const(), phase_ansatz=ro.pulses.SinCrab(2))

# Initial pulse parameter guess
initial_params = ro.pulses.PulseParams(duration=7.0, detuning_params=[0.0], phase_params=[0.0, 0.0])

# Optimize the pulse parameters
opt_result = ro.optimization.optimize(gate, pulse_ansatz, initial_params, tol=1e-10)
optimized_params = opt_result.params

# Plot the pulse
ro.characterization.plot_pulse(pulse_ansatz, optimized_params)

Citing RydOpt

If you find this library useful for your research, please cite:

D.F. Locher, J. Old, K. Brechtelsbauer, J. Holschbach, H.P. Büchler, S. Weber, M. Müller, Multiqubit Rydberg Gates for Quantum Error Correction, PRX Quantum 7, 020354 (2026)

Contributors

The following people have, so far, contributed to the development of RydOpt:

• David Locher
• Sebastian Weber
• Jakob Holschbach
• Javad Kazemi

We warmly welcome new contributions! Please refer to the contributor guide for more information!

The development of RydOpt has been supported by Forschungszentrum Jülich, RWTH Aachen University, University of Stuttgart, and the company ParityQC. We acknowledge support from the Federal Ministry of Research, Technology and Space (BMFTR) through the grant MUNIQC-Atoms and from the German Research Foundation (DFG) through the priority programme SPP 2514.

License

The RydOpt software is licensed under the MIT License.