Jiuzhang 4.0: China’s Quantum Leap to 3,050 Photons
In a landmark achievement published in the journal Nature on May 13, 2026, Chinese scientists unveiled “Jiuzhang 4.0,” a programmable photonic quantum computing prototype that manipulates and detects up to 3,050 photons — a tenfold increase over the previous record of 255 photons set by Jiuzhang 3.0 in 2023. The system demonstrated a quantum speedup exceeding 10^54 times over the world’s most powerful classical supercomputer for the specific task of Gaussian boson sampling (GBS), according to Xinhua News.
But behind the staggering numbers lies an equally compelling story: the breakthrough was driven by a team of six PhD students with an average age of just 24, whose journey from experimental novices to lead authors embodies China’s strategy of cultivating a new generation of quantum scientists.
The Jiuzhang Lineage: From 76 to 3,050 Photons
Named after the ancient Chinese mathematical classic Jiuzhang Suanshu (“The Nine Chapters on the Mathematical Art”), the Jiuzhang series of photonic quantum computers has undergone rapid evolution since its debut in December 2020. Jiuzhang 1.0 made history as the first photonic quantum computer to claim quantum supremacy, solving a GBS problem in 200 seconds that would have taken the Sunway TaihuLight supercomputer 2.5 billion years. Jiuzhang 2.0 followed in 2021 with improved photon handling, and Jiuzhang 3.0 in 2023 achieved a 255-photon milestone with a 10-quadrillion-fold speedup.
Jiuzhang 4.0 represents a quantum leap — literally and figuratively. As CGTN reported, the new prototype generates a sample in just 25.6 microseconds, while the world’s fastest supercomputer, El Capitan, would require more than 10^42 years to produce the equivalent result.
The Technical Breakthrough: Adding Time to Space
The key innovation behind Jiuzhang 4.0 is a programmable spatiotemporal hybrid-coded architecture. Traditional photonic quantum computers use “pure spatial encoding” — adding more photons requires proportionally more optical components, leading to photon loss that limits scalability. The USTC team, led by academician Pan Jianwei and professor Lu Chaoyang, solved this by adding a time dimension to the encoding.
As detailed by the State Council Information Office, the system uses a cascaded architecture of three 16-mode interferometers connected by fiber delay loops, achieving cubic connectivity scaling (16^3 = 4,096 modes) with only linearly growing physical resources. The team integrated 1,024 high-efficiency squeezed-state optical fields into an 8,176-mode circuit, achieving 92% source efficiency and 51% overall system efficiency.
The Light of Youth: A Team of 24-Year-Olds
The human story behind Jiuzhang 4.0 is as remarkable as the technical achievement. A feature article published by Xinhua on June 1, 2026, titled “The Light of Youth in Jiuzhang,” profiles six PhD students who formed the core technical team.
Liu Hualiang, the first author of the Jiuzhang 4.0 paper, joined USTC at age 17 in 2016. He progressed from “experimental novice” on Jiuzhang 2.0 to core member on Jiuzhang 3.0 and finally lead author on Jiuzhang 4.0. He designed the unique three-stage cascaded spatiotemporal hybrid-coded interferometer — the “engine” of the new system. “We spent three consecutive Spring Festivals in the lab,” Liu recalled. “On the 2024 New Year’s Eve, the entire campus was empty, only our group gathered around the lab eating dumplings while staring at experimental data on the screens.”
Gong Siqiu, born in 1999, self-recommended to Professor Lu Chaoyang as a freshman. His journey mirrors the Jiuzhang evolution itself: from “acknowledgments” in Jiuzhang 1.0 to co-first author on Jiuzhang 3.0 and 4.0. He was responsible for matrix measurement and algorithm optimization, reducing computational elements from approximately 10 million to under 1,000 through clever analysis.
Su Hao, born after 2000, started as an assistant handling data and moving equipment before becoming the team’s “construction expert.” He built and rebuilt the three complex interferometers over 10 iterations to achieve micron-level precision.
“Approaching the light, following the light, becoming the light, emitting the light — we have fully experienced this process,” Liu Hualiang said.
Expert Analysis: What Jiuzhang 4.0 Really Means
Independent analysis by Marin Ivezic, founder of Applied Quantum and author of PostQuantum.com, provides important context. Ivezic confirms that the engineering is real and significant — the 92% source efficiency and cascaded architecture are genuine advances, and the validation methodology is thorough, with raw data made publicly available.
However, Ivezic cautions against conflating this specialized achievement with general-purpose quantum computing. Gaussian boson sampling is an “intermediate model” that does not implement universal gate operations, error correction, or run Shor’s algorithm. The result does not directly advance the path to a cryptographically relevant quantum computer (CRQC).
“The right frame,” Ivezic writes, “is that Jiuzhang 4.0 is a significant photonic engineering achievement that strengthens China’s position in one branch of the quantum technology tree. For readers focused on cryptographic security timelines, it changes nothing.”
What’s Next
The Jiuzhang 4.0 architecture opens a plausible path toward “trillion-qumode three-dimensional cluster states” — a potential foundation for fault-tolerant photonic quantum computing. This positions USTC as a competitive player alongside PsiQuantum (US) and Xanadu (Canada) in the race toward practical photonic quantum computers.
For now, the achievement stands as a testament to both Chinese scientific ambition and the power of investing in young researchers. As Gong Siqiu wrote in his thesis acknowledgments: “Light races through wondrous realms opening new horizons; calculations break through the highest heavens, surpassing distant extremes.”