Co-Optimization and Error Characterization of Single- and Two-Qubit Gates in a superconducting quantum device
ORAL
Abstract
Simultaneously achieving high-fidelity single- and two-qubit gates remains a key challenge in superconducting quantum processors, as device parameters that improve one operation often degrade the other. We demonstrate that careful co-optimization of qubit–coupler coupling strengths in a two-transmon circuit with a tunable coupler enables high-fidelity performance across both gate types without compromising readout. Using a diabatic CZ gate calibrated with the Phase-Averaged Leakage Error Amplification (PALEA) protocol, we isolate and suppress leakage processes while remaining insensitive to coherent phase errors. Iterative interleaved randomized benchmarking yields a 40-h-averaged CZ fidelity of 99.93%, alongside simultaneous single-qubit fidelities of 99.98% [1]. A detailed error budget resolves contributions from f-state leakage, coupler leakage, and SWAP errors, confirming incoherent processes as the dominant limitation. Complementary measurements show >99.94 % readout fidelity, underscoring the effectiveness of the system-level optimization. Together, these results highlight how calibrated coupling design and targeted leakage suppression enable consistently high-fidelity operation across all gate primitives in a single superconducting device.
[1] Marxer et al., arXiv preprint arXiv:2508.16437 (2025)
[1] Marxer et al., arXiv preprint arXiv:2508.16437 (2025)
–
Publication: Marxer et al., arXiv preprint arXiv:2508.16437 (2025)
Presenters
-
Fabian Marxer
- IQM Quantum Computers