China’s latest quantum computers, Zuchongzhi 3.0 and Jiuzhang 4.0, have achieved massive quantum advantage. The 105-qubit superconducting Zuchongzhi 3.0 finished a random circuit sampling task in seconds that would take the fastest classical supercomputers billions of years, outperforming Google’s latest chips by millions of times. Jiuzhang 4.0, a photonic system using over 2,000 photons, solved a Gaussian boson sampling problem in minutes that no classical machine could finish in realistic time. Both systems show extremely high gate fidelity, long coherence times, and advanced error mitigation. While real-world applications in drug discovery, optimization, and cryptography are still years away, these breakthroughs push quantum technology far beyond classical limits and highlight China’s lead in the global quantum race.
Long Version
China’s Quantum Leap: Revolutionizing Computing with Unprecedented Speed and Power
In the rapidly evolving landscape of quantum computing, China has once again demonstrated its prowess with the deployment of a groundbreaking quantum computer that solves complex computational tasks in mere minutes—challenges that would demand billions of years from even the world’s fastest supercomputers like Frontier or Sunway TaihuLight. This achievement not only reaffirms quantum advantage but also edges closer to quantum supremacy, where quantum processors far outstrip classical computers in performance for specific problems. As of November 2025, the latest milestone involves the Tianyan-287, a superconducting quantum computer housed in Hefei, which leverages advanced quantum chips to deliver calculations 450 million times faster than leading classical systems. This huge leap forward underscores how these machines are pushing beyond classical limits, though practical real-world applications remain years away.
Understanding the Foundations: From Classical to Quantum Computing
At the heart of this breakthrough lies the fundamental shift from classical computers, which process information in binary bits, to quantum computers that harness qubits. Unlike traditional bits, qubits exploit principles like superposition—allowing them to exist in multiple states simultaneously—and entanglement, where the state of one qubit instantly influences another, regardless of distance. These properties enable quantum processors to tackle intricate calculations exponentially faster.
China’s advancements span two primary routes: superconducting qubits, as seen in the Zuchongzhi series, and photonic quantum computers like Jiuzhang. Superconducting qubits, often transmon qubits, rely on Josephson junctions and operate at near-absolute zero temperatures to maintain coherence. In contrast, photonic systems use photons manipulated through optical circuits, beam splitters, and mirrors to perform tasks like boson sampling. Both approaches demand high fidelity in operations, including single-qubit gates and two-qubit gates, to minimize errors and sustain quantum states over extended coherence times.
The Zuchongzhi 3.0: A Superconducting Powerhouse
Launched in March 2025 by researchers at the University of Science and Technology of China (USTC) and the Chinese Academy of Sciences (CAS), the Zuchongzhi 3.0 represents a pinnacle in superconducting quantum computing. This 105-qubit quantum processor features 182 couplers in a 2D grid architecture, enhancing interconnectivity and scaling potential. Built using advanced fabrication techniques like the indium bump flip-chip process, it achieves remarkable metrics: a coherence time of 72 microseconds, relaxation time (T1) and dephasing time (T2) optimized for stability, parallel single-qubit gate fidelity of 99.90%, two-qubit gate fidelity of 99.62%, and readout fidelity of 99.13%.
The system’s benchmark experiment involved random circuit sampling (RCS), a key test for quantum advantage. Using 83 qubits across 32 cycles, Zuchongzhi 3.0 completed the task in seconds—a feat estimated to take the world’s most powerful supercomputer 6.4 billion years. This delivers a speedup of 10^15 times over classical benchmarks like tensor network algorithms, and a million times faster than Google’s Sycamore processor. Compared to Google’s Willow, which also claims quantum advantage, Zuchongzhi 3.0 sets a new record in superconducting systems by outperforming it by six orders of magnitude.
The research, led by Pan Jianwei, Zhu Xiaobo, and Peng Chengzhi at the Hefei National Laboratory, was published in Physical Review Letters and pre-released on arXiv. It builds on predecessors like Zuchongzhi 2.1, a 66-qubit system that first achieved quantum advantage in superconducting routes in 2021. Error correction remains a focus, with plans for surface codes at distances 7, 9, and 11 to combat noise and improve output distribution accuracy. In a significant step toward commercialization, China opened this superconducting quantum computer for commercial use in October 2025, allowing businesses and researchers to access its capabilities for practical experiments and development.
Jiuzhang 4.0: Pushing Photonic Boundaries
Complementing the superconducting track, China’s photonic quantum computer lineage shines with Jiuzhang 4.0, unveiled in August 2025 by USTC scientists. This system integrates over 2,000 photons, far surpassing Jiuzhang 3.0’s 255 photons from 2023. It excels in Gaussian boson sampling, a computational task where photons navigate an optical circuit to produce probability distributions infeasible for classical simulation.
Jiuzhang 4.0 solved its benchmark problem in minutes, a calculation that would require the U.S. supercomputer El Capitan—a successor to Frontier—a million trillion trillion trillion years. This demonstrates quantum supremacy in photonic domains, leveraging high gate fidelity and minimal loss in photon manipulation. Earlier iterations, like the original Jiuzhang with 76 photons and Jiuzhang 2.0 with 113, laid the groundwork by achieving quantum advantage in 2020 and 2021, respectively. Recent enhancements in integrated photonic quantum chips, achieved in February 2025, have further boosted multipartite entanglement and continuous-variable quantum processing, enabling more robust simulations for chemistry and materials science.
Emerging Innovations: Optical Quantum Chips and Hybrid Systems
Beyond these prototypes, China is advancing hybrid technologies. In November 2025, the Chip Hub for Integrated Photonics Xplore (CHIPX) introduced a scalable optical quantum chip with over 1,000 components on a 6-inch silicon wafer. This monolithic design processes AI workloads 1,000 times faster than Nvidia GPUs, with production ramping to 12,000 wafers annually. While yields are low, its integration of photons and electronics hints at practical applications in aerospace and finance, bridging quantum and classical computing.
Additionally, breakthroughs in analog computing complement quantum efforts. In October 2025, researchers from Peking University developed a resistive random-access memory chip capable of speeds 1,000 times faster than high-end Nvidia GPUs, solving century-old problems in complex simulations. Though not purely quantum, this analog approach integrates with quantum systems for hybrid modeling, as seen in the Sunway supercomputer’s use of 37 million processor cores to simulate quantum chemistry at molecular scales, fusing AI and quantum science for unprecedented accuracy.
Institutions and Collaborative Ecosystem
These breakthroughs stem from a robust ecosystem involving USTC, CAS, Hefei National Laboratory, and companies like QuantumCTek Co., Ltd., CIQTEK, TuringQ, and QBoson. QuantumCTek, for instance, has built a comprehensive supply chain from research to manufacturing, accelerating tech transfer. The Tianyan-287, deployed by China Telecom Quantum Group, connects to a global cloud platform, fostering international collaboration with over 37 million accesses from 60+ countries.
Challenges and the Path to Practical Applications
Despite the excitement, hurdles persist. Maintaining high fidelity amid scaling requires advanced error correction to handle noise from relaxation and dephasing. Current systems excel in narrow experiments but struggle with general-purpose tasks. Real-world applications—in quantum chemistry for drug discovery, complex algorithm optimization, or building a quantum internet—are still emerging. Examples include quantum-encrypted power grids, faster autonomous parking simulations, and precision sensing for batteries.
China’s five-year plan prioritizes quantum alongside bio-manufacturing and 6G, spurring start-ups and domestic production of components like dilution refrigerators. As studies continue in quantum simulation and entanglement, these efforts position China as a leader, reshaping global computing dynamics amid intensifying U.S.-China competition in quantum technologies.
This progress, while transformative, demands ongoing research to bridge the gap to everyday use. For now, it stands as a testament to human ingenuity, proving that quantum machines are not just theoretical—they’re redefining what’s computationally possible.
