The Zuchongzhi-3, a superconducting quantum computing prototype boasting 105 qubits and 182 couplers, has revolutionized random quantum circuit sampling. Developed by a team at the University of Science and Technology of China (USTC), this prototype operates at a speed 10^15 times faster than the fastest supercomputer available and a million times faster than Google’s latest results.
This achievement is a landmark in quantum computation, following the success of Zuchongzhi-2. The research has been highlighted as the cover article in Physical Review Letters.
Quantum Supremacy: Redefining Computational Boundaries
Quantum supremacy signifies the ability of quantum computers to perform tasks that classical computers cannot. In 2019, Google’s 53-qubit Sycamore processor completed a random circuit sampling task in 200 seconds, a task that would have taken about 10,000 years on the fastest supercomputer of that time.
However, in 2023, USTC researchers used advanced classical algorithms to complete the same task in just 14 seconds with over 1,400 A100 GPUs. Utilizing Frontier supercomputers, the task time was reduced to 1.6 seconds, challenging Google’s claim of quantum supremacy.
Advancements and Key Metrics
Building on the 66-qubit Zuchongzhi-2, the USTC team enhanced key performance metrics to create Zuchongzhi-3. This quantum processor features 105 qubits and 182 couplers, achieving a coherence time of 72 μs, parallel single-qubit gate fidelity of 99.90%, parallel two-qubit gate fidelity of 99.62%, and parallel readout fidelity of 99.13%.
The extended coherence time allows for more complex operations and computations. In an 83-qubit, 32-layer random circuit sampling task, Zuchongzhi-3’s computational speed surpassed the world’s most powerful supercomputer by 15 orders of magnitude, outperforming Google’s October 2024 results by 6 orders of magnitude.
Future Directions: Research and Development
Following the achievement with Zuchongzhi-3, the team is advancing research in quantum error correction, quantum entanglement, quantum simulation, and quantum chemistry. They adopted a 2D grid qubit architecture for efficient interconnections among qubits and enhanced data transfer rates.
The team integrated a surface code and is actively researching quantum error correction with a distance-7 surface code, with plans to increase this distance to 9 and 11. This sets the stage for massive integration and manipulation of quantum bits.
Collaborative Efforts and Recognition
The significance of the team’s work has received widespread acclaim. A journal reviewer described it as “benchmarking a new superconducting quantum computer with state-of-the-art performance” and a “significant upgrade from the previous 66-qubit device (Zuchongzhi-2).”
The research team includes Pan Jianwei, Zhu Xiaobo, and Peng Chengzhi, in collaboration with the Shanghai Research Center for Quantum Sciences, Henan Key Laboratory of Quantum Information and Cryptography, China National Institute of Metrology, Jinan Institute of Quantum Technology, School of Microelectronics at Xidian University, and the Institute of Theoretical Physics under the Chinese Academy of Sciences.
Reference: Dongxin Gao et al, Establishing a New Benchmark in Quantum Computational Advantage with 105-qubit Zuchongzhi 3.0 Processor, Physical Review Letters (2025). DOI: 10.1103/PhysRevLett.134.090601
