When the Quantum World Learns to Think in Parallel: A Journey into Quantum Computing and the Zuchongzhi 3.0 Processor

Forum topic · 小凯 · 2026-03-02

Summary

This Chinese forum post offers a popular-science explainer on quantum computing, centered on two landmark 105-qubit milestones: Google's Willow chip (published in Nature, December 2024), which achieved below-threshold quantum error correction with error rates falling exponentially as qubit arrays scaled from 3x3 to 7x7, and China's Zuchongzhi 3.0 superconducting processor (from USTC's Pan Jianwei team, announced March 2025 in Physical Review Letters), which completed 83-qubit, 32-cycle random circuit sampling quadrillions of times faster than the fastest classical supercomputers and a million times faster than Google's earlier Sycamore chip. The article explains core concepts—qubits, superposition (n qubits representing 2^n states), and entanglement—in layperson terms, compares the two research routes (error-correction-first versus scale-first), and surveys applications including drug discovery, materials science, financial modeling, quantum machine learning, and cryptography (Shor's algorithm versus quantum key distribution). It outlines the roadmap from quantum supremacy to quantum simulators and universal fault-tolerant quantum computers, with global competition context spanning the US, China, Canada, and Europe, plus NVIDIA's 2025 Quantum Day.

When the Quantum World Learns to Think in Parallel: A Journey into Quantum Computing

> Imagine walking every path of a maze simultaneously instead of trying them one by one. That is what quantum computers do every day. In early 2025, a lab in Hefei, China unveiled Zuchongzhi 3.0, a 105-qubit superconducting quantum computer that can finish in seconds tasks that would take classical supercomputers quadrillions of years—far longer than the age of the universe.

Chapter 1: From Switches to Spirits — The Magic of Qubits

Classical computers' limited approach

A classical bit is like a light switch: either 0 or 1, never anything else. Everything from phones to the internet is built from countless such switches. But for problems like simulating drug molecules binding to viruses, classical computers must try paths one at a time—like searching an unnumbered book in a library of one million volumes.

Qubits: superposition in the microscopic world

In the quantum world, particles can exist in superposition—like a coin that is simultaneously heads and tails until you grab it. A qubit can be 0 and 1 at once; two qubits represent 00, 01, 10, and 11 simultaneously:

n \text{ qubits} \rightarrow 2^n \text{ simultaneous states}

Zuchongzhi 3.0's 105 qubits handle $2^{105}$ states—a 32-digit number exceeding the count of all sand grains on Earth.

Entanglement: quantum "telepathy"

Einstein called quantum entanglement "spooky action at a distance." Measuring one entangled qubit instantly reveals the state of its partner, no matter the distance. Experiments have repeatedly confirmed entanglement is real, and it underpins quantum communication and theoretically unbreakable encryption.

Chapter 2: The Duel — Zuchongzhi 3.0 vs. Willow

December 2024 was historic: within one week, two 105-qubit results appeared.

Dec 9, 2024: Google published Willow in *Nature*, solving a ~30-year-old problem in quantum error correction.
Dec 17, 2024: USTC's Pan Jianwei team released Zuchongzhi 3.0 results on arXiv (later *Physical Review Letters*).

Willow's breakthrough: error correction that improves with scale

Qubits are fragile, operating near absolute zero (~-273°C); more qubits traditionally meant more errors. Willow achieved below-threshold error correction: scaling the array from 3×3 to 5×5 to 7×7 caused logical error rates to fall exponentially—by a factor of about 2.14 per step. The logical qubit's lifetime (291 µs) exceeded its best physical qubit's lifetime (119 µs) by 2.4×, meaning corrected quantum information outlives the raw version—a key step toward practical quantum computing.

Zuchongzhi 3.0: pushing quantum advantage

| Metric | Zuchongzhi 2.0 (2021) | Zuchongzhi 3.0 (2025) | |---|---|---| | Readable qubits | 66 | 105 (+59%) | | Coupled qubits | – | 182 | | Single-qubit gate fidelity | 99.7% | 99.90% | | Two-qubit gate fidelity | 99.2% | 99.62% | | Readout fidelity | – | 99.13% | | Coherence time | – | 72 µs |

In the random circuit sampling benchmark, Zuchongzhi 3.0 completed an 83-qubit, 32-cycle sampling task quadrillions of times ( $10^{15}$ ) faster than the world's fastest supercomputer—so a 1,000-year classical calculation takes under one second—and was a million times faster than Google's 67-qubit Sycamore result of October 2024.

Two routes, one dream

Willow: error-correction-first — proving reliability can scale, the necessary path to universal quantum computing.
Zuchongzhi 3.0: scale-first — exploring the ceiling of superconducting quantum advantage with current technology.

The routes are complementary: per USTC professor Zhu Xiaobo, Zuchongzhi 3.0 lays the foundation for future quantum error correction and universal quantum computing.

Chapter 3: What Can Quantum Computing Do?

Today's quantum computers resemble early mainframes—huge, specialized, not desktop machines. But like ENIAC (30 tons, 170 m², 5,000 additions/second in 1946), they will evolve rapidly. Promising applications:

1. Drug discovery — simulating molecular–protein interactions at the atomic level, where classical simulation is impractical. Unlike AI tools such as AlphaFold (structure prediction), quantum computers can simulate dynamics. 2. Materials science — unlocking the mechanism of high-temperature superconductivity, a quantum many-body mystery beyond classical simulation. 3. Financial modeling — portfolio optimization, risk analysis, and derivatives pricing; firms like JPMorgan and Goldman Sachs are already exploring quantum algorithms. 4. Artificial intelligence — quantum machine learning (QML) may speed up high-dimensional AI training by orders of magnitude; hybrid quantum-classical architectures may dominate. 5. Cryptography — Shor's algorithm threatens RSA encryption, while quantum key distribution offers theoretically unbreakable communication; China operates a nationwide quantum communication network.

Chapter 4: What Comes Next

Three stages of development:

1. Quantum supremacy — achieved by Google's Sycamore and China's Jiuzhang and Zuchongzhi series. 2. Quantum simulators — hundreds to thousands of qubits for specialized scientific problems; we are at this threshold. 3. Universal quantum computers — programmable, error-corrected machines running arbitrary quantum algorithms, possibly 10–15 years away.

China's roadmap: beyond Zuchongzhi 3.0's 105 qubits, the team plans surface-code error-correction experiments at code distances 7, 9, and 11. On the photonic route, Jiuzhang 3.0 manipulates 255 photons, a hundred quadrillion times faster than supercomputers on specific problems.

Global landscape: US (Google, IBM, Microsoft), China (USTC, strong on both superconducting and photonic routes), Canada (D-Wave annealing), and Europe (ion-trap advances such as IonQ). In March 2025, NVIDIA held its first Quantum Day and founded the NVIDIA Accelerated Quantum Research Center (NVAQC); CEO Jensen Huang revised his earlier 20-year commercialization forecast, saying quantum computing "is approaching an inflection point."

Epilogue: On the Eve of a Revolution

Quantum computers will not replace laptops—they are specialized tools for problems classical machines cannot touch. But the impact could be profound: drug development shortened from 10 years to 1, breakthroughs in energy materials, leaps in AI capability, unbreakable privacy. Like the transistor in 1947, today's quantum milestones are the opening chapter of a revolution.

References

1. Gao, D., Fan, D., Zha, C., et al. (2025). "Establishing a New Benchmark in Quantum Computational Advantage with 105-qubit Zuchongzhi 3.0 Processor." *Physical Review Letters*, 134, 090601. https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.134.090601 2. Google Quantum AI. (2024). "Quantum error correction below the surface code threshold." *Nature*, 638, 920-926. https://www.nature.com/articles/s41586-024-08449-y 3. Pan, J.W., et al. (2021). "Quantum computational advantage using photonic Gaussian boson sampling." *Physical Review Letters*, 127, 180502. 4. University of Science and Technology of China (2025). Announcement of the Zuchongzhi 3.0 superconducting quantum prototype. https://news.ustc.edu.cn/info/1056/90742.htm 5. Chinese Academy of Sciences. "Top 10 Chinese science and technology news of 2025." http://www.cas.cn/cm/202601/t20260130_5099165.shtml

*Written in a Feynman-style popular-science approach; corrections on scientific points are welcome.*