TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The University of Science and Technology of China’s Jiuzhang 4.0 was among 20 projects honored at the 2026 World Internet Conference Leading Science and Technology Awards. Its reported photon-counting results mark progress in a contested quantum-computing benchmark, but Gaussian boson sampling has no known commercial use and the speed comparison remains tied to a specialized task.
China’s Jiuzhang 4.0 photonic quantum computer, developed by researchers at the University of Science and Technology of China (USTC), was one of 20 projects honored at the 2026 World Internet Conference Leading Science and Technology Awards in Wuzhen. The award recognizes the research amid continuing debate over how far its specialized quantum benchmark shows an advantage over conventional computers.
The awards were presented Monday in Wuzhen, in eastern China’s Zhejiang province, according to state broadcaster CGTN, as reported by The Quantum Insider. Organizers received more than 430 submissions from 43 countries and regions. The 20 winning projects were divided among basic research, key technologies, and engineering and development: two, 10, and eight awards, respectively. Other recognized work included optical transmission technology for space-based quantum information and an artificial-intelligence agent platform for intelligent vehicles.
Jiuzhang processes information using photons, or particles of light, directed through an optical circuit and counted by detectors. In work described by the team in Nature in May, researchers sent 1,024 beams of squeezed light through a circuit with 8,176 modes, or channels. The device detected up to 3,050 photons in a single run, compared with 255 for Jiuzhang 3.0. It performs Gaussian boson sampling, a task that measures patterns of photons emerging from the optical network.
USTC professor Lu Chaoyang said the system produced its most complex sample in about 25 microseconds. Xinhua quoted Lu as saying that calculating the same result would take the world’s most powerful supercomputer more than 1042 years. The research team’s comparison, described in its preprint, names the classical system as El Capitan and estimates a speed difference exceeding 1054. Those figures concern the specific sampling task, not a general-purpose computing workload.
What the Award Recognizes
The recognition gives public and institutional visibility to China’s photonic quantum-computing research and places Jiuzhang among projects selected from a large international submission pool. Its technical results also add to a central question in quantum computing: whether a quantum device can perform a carefully chosen task that is prohibitively difficult for classical machines.
That question matters for evaluating the field, but the result should not be confused with a demonstration that Jiuzhang can outperform conventional computers on everyday applications. Gaussian boson sampling is a specialized benchmark, and the source report says it has no known commercial use. The award recognizes the project; it does not by itself resolve disputes about the benchmark or establish a practical quantum computer.
The reported photon count is a concrete measure of experimental scale: Jiuzhang 4.0 detected substantially more photons in a run than its predecessor. The broader significance depends on whether the platform’s methods can support more useful computing, including approaches to error correction, rather than only sampling demonstrations.
Products Worth Considering
Apple MacBook Air 13-inch with M5 chip, 16GB memory, 512GB SSD, and 12MP Center Stage camera
Anker SOLIX C300 Portable Power Station, 288Wh
Samsung Galaxy Tab S10 Lite – 6 GB RAM, 128 GB storage, up to 2 TB microSD, 10.9‑inch LCD, Gray
As an affiliate, we earn on qualifying purchases.
From Jiuzhang 1.0 to 4.0
The Jiuzhang program comes from the research group led by physicist Pan Jianwei at USTC. The series began in 2020 with a machine that detected 76 photons. Jiuzhang 2.0 reached 113, and the 3.0 system recorded up to 255, according to prior reporting cited by The Quantum Insider. The 4.0 result is the latest step in a program focused on optical quantum devices.
Photonic systems differ from quantum computers built around superconducting circuits or trapped atoms: they use light and optical components to carry out the computation. The Jiuzhang team’s work was described in Nature in May, while a separate paper assessing its latest claims was posted as a preprint on arXiv, according to the source report. A preprint has not necessarily undergone peer review, so its findings and comparisons should be treated accordingly.
Earlier boson-sampling claims have faced a challenge from a 2024 Nature Physics study by researchers at the University of Chicago and Argonne National Laboratory. That study used photon loss—a common experimental limitation—to simulate earlier sampling results on conventional computers. Its authors said their classical method matched ideal results better than the experiments did. The Jiuzhang 4.0 team says its newer work was designed to address that approach.
“The device produces its most complex sample in about 25 microseconds, “shorter than the blink of an eye.””
— Lu Chaoyang, USTC professor, as quoted by Xinhua
Limits of the Speed Comparison
The award announcement does not settle how robust Jiuzhang 4.0’s claimed advantage is under different assumptions or against future classical algorithms. The source report says the team’s preprint found that the machine’s outputs could be distinguished from several classical imitations, but the full scope of that comparison and its independent evaluation are not established in the supplied material. The reported time ratios are estimates, not measurements of a broad range of useful workloads.
It is also unclear from the available reporting whether the arXiv study has completed peer review, what additional validation will be published, or whether the hardware can be scaled while maintaining the reported performance. The team has discussed a route toward error-corrected optical computing, but the award and sampling result do not show that such a system has been built. No commercial application for Gaussian boson sampling is identified.
Research Beyond the Sampling Test
The next evidence to watch for is further analysis of the Jiuzhang 4.0 results, including peer-reviewed publication or independent assessment of the team’s comparison with classical simulations. Such work could clarify how well the experiment addresses the photon-loss challenge and how its performance changes as system scale and experimental conditions vary.
Lu said the team’s light sources and circuit design could support the creation of large three-dimensional cluster states—networks of linked light that researchers view as a possible foundation for optical quantum computers capable of correcting their own errors. That is a research direction, not a demonstrated capability of the award-winning prototype. The available report gives no schedule for a follow-up system or commercial deployment, so those milestones remain unknown.
Key Questions
What did Jiuzhang 4.0 win?
It was one of 20 projects honored with the 2026 World Internet Conference Leading Science and Technology Awards at a ceremony in Wuzhen, China.
What does Jiuzhang 4.0 do?
The USTC prototype uses photons passing through an optical circuit to perform Gaussian boson sampling, a specialized task that records patterns of photons detected after they pass through the system.
How many photons did it detect?
The team reported that Jiuzhang 4.0 detected up to 3,050 photons in a single run. The source report gives 255 as the comparable figure for Jiuzhang 3.0.
Does the award prove Jiuzhang is faster for practical computing?
No. The speed claims relate to a specific sampling benchmark. The source report says Gaussian boson sampling has no known commercial use, and earlier advantage claims have faced challenges from classical simulations.
What remains uncertain about the research?
The available reporting does not establish whether the latest preprint has completed peer review or how the claimed advantage will hold up under independent testing and future classical methods. A path to error-corrected optical computing is described as a research goal, not an achieved result.
Source: thequantuminsider.com
