A Chinese research team has broken a long-standing bottleneck in superconducting quantum computing, managing to keep two-qubit gates fast without sacrificing accuracy.

The team, jointly formed by Origin Quantum and the University of Science and Technology of China, proposed a scheme called the "parameter-space expansion controlled-Z (PSE-CZ) gate." The work has been published in the international physics journal Physical Review Letters, with experiments carried out on China's self-developed superconducting quantum computer "Origin Wukong."

Quantum gates are the basic operations behind any quantum computation. For reliable results, they must combine ultra-high speed with ultra-high fidelity. For years, there was a trade-off: faster gates worsen waveform distortion and timing errors, hurting precision, while slower operations drag down overall performance.

To address this, the team tested the PSE-CZ scheme on 20 pairs of two-qubit gates using Origin Wukong. Results showed the approach suppresses errors caused by short-time distortion, pushing gate performance closer to the dephasing limit. Even at extremely short gate times of 30 to 40 nanoseconds, PSE-CZ outperformed conventional CZ gates.

The scheme is not limited to superconducting systems. It could extend to other platforms such as ion traps and solid-state spin qubits, promising faster and higher-fidelity quantum logic operations.