36Kr Exclusive | A team from Jiaotong University has secured a 100-million-yuan financing round led by China Internet Investment Fund and Tsinghua-affiliated capital, stepping into the quantum computing sector with the "atom + photonic chip" technology.
Image source / Enterprise
This article is approximately 3400 words long, with an estimated reading time of 8 minutes
Author | Ou Xue
Editor | Yuan Silai
Hard Krypton learned that Atomology, a neutral-atom quantum computing enterprise, has recently completed a 100-million-yuan-level angel round financing. This round of financing is jointly led by China Internet Investment Fund and Tsinghua-affiliated capital Muhua Tech Venture Capital, with nearly 10 other institutions participating in follow-on investment. The funds will be mainly used for core technology R&D, engineering platform construction and core team expansion, to accelerate the evolution of neutral-atom quantum computing systems from laboratory prototypes to engineered, large-scale and industrial-grade products.
Founded in April 2026, Atomology has chosen a path different from traditional scientific research teams since its inception: large-scale engineerable quantum computing based on neutral atoms and photonic chips.
At present, R&D personnel account for more than 80% of the company's total headcount, and personnel with master's or doctoral degrees account for over 90%. Most of the core members come from top universities such as Shanghai Jiao Tong University, University of Science and Technology of China, and the University of Oxford. It is worth noting that Atomology was incubated by Turing Quantum, a domestic optical quantum computing enterprise. Its core team has been deeply engaged in the Integrated Quantum Information Technology Research Center (IQIT) of Shanghai Jiao Tong University for many years, and has fully inherited the technical accumulation in the fields of cold atoms, photonic chips and metasurface optics, which provides underlying support for the team to rapidly promote engineering implementation.
The development of quantum computing is essentially an exploration to find the optimal qubit carrier. Neutral atoms are recognized as one of the most promising routes due to their natural indistinguishability, long coherence time and large-scale scaling potential brought by optical tweezer arrays.
However, the industry generally faces a harsh reality: technological breakthroughs do not equal industrial breakthroughs. At present, most neutral-atom systems are still composed of hundreds of free-space optical components, huge lasers and complex control systems, which are essentially "laboratory devices" that highly rely on manual debugging. How to achieve large-scale manufacturing and long-term stable operation has become the core pain point restricting industrialization.
Fang Xiaotian, COO of Atomology, has a vivid metaphor for this: "The traditional experimental platform is like a precision instrument that requires careful maintenance by experienced operators. Once the environment fluctuates, more than 50% of the time may be spent on recovery work."
In the face of this persistent industry problem, Atomology's solution is to advance both "chip-based design" and "modularization" at the same time, trying to turn quantum computers from "hand-polished products" into "assembly-line manufactured products".
In terms of chip-based design, the team has focused on overcoming the physical limit of traditional optical tweezer technology that is difficult to scale up on a large scale. Traditional solutions rely on dynamic optical devices such as AOD and SLM, which not only have a scale upper limit locked at the order of 10,000, but also have serious mixing crosstalk and algorithm delay.
Atomology focuses on wafer-level ultra-precision metasurface photonic chips, and uses metasurface technology to generate millions of high-density static optical tweezer arrays on a single chip at one time. The optical path is solidified without complex spatial optical path debugging. At the same time, the team combines on-chip modulators (such as lithium niobate electro-optic modulators) with femtosecond laser direct writing 3D optical waveguides to realize full-chip integration of optical paths and modulators, completely solving the pain points of dynamic optical tweezers such as "insufficient resolution, site crosstalk, and large delay". At present, the bandwidth of its on-chip regulation chip has exceeded 110GHz, and the iteration cycle of high-density optical waveguides has been shortened to 3 days.
In terms of modularization, for peripheral devices that cannot be chip-based for the time being, the team converts them into standardized engineering modules with high precision and high stability. Fang Xiaotian emphasized that the academic circle pursues the highest indicator of "doing it well once", while the industrial circle requires stable and reproducible output of "doing it well at one time".
Through active feedback control, constant temperature system and precision packaging technology, Atomology converts the complex macroscopic optical path system into a pluggable and upgradable industrial-grade product, which endows it with the capabilities of assembly line manufacturing, rapid iteration, easy deployment and maintenance.
This "chip + module" engineering route has also been verified by the industry. In July 2026, 31 top global institutions and enterprises including MIT, Harvard, QuEra and Pasqal jointly published a paper on neutral-atom quantum computing strategy, which dedicated a chapter to discussing the importance of "integrated photonic chips" for future large-scale scaling, covering directions such as metasurface optical tweezers, on-chip regulation and high-density optical waveguides. This is highly consistent with Atomology's previous technical layout, which the team describes as "a feeling of finally being verified".
In terms of technology evolution route, Atomology has a clear three-step plan: large-scale static optical tweezer array, reconfigurable optical tweezer array, high-fidelity manipulation and storage. Centering on these three core points, the team has formulated clear R&D OKRs: in terms of scale, realize atomic arrays of 100,000 to 1,000,000 levels with the help of metasurface chips, and advance toward the million-qubit super quantum computing platform; in terms of reconfigurability, achieve high-speed real-time feedback control with bandwidth greater than 200MHz and optical tweezer reconfiguration time less than 50 milliseconds; in terms of fidelity, combine low-temperature environment and ultra-high vacuum technology to increase the atomic lifetime to the order of thousands of seconds, with the target single-qubit gate fidelity greater than 99.99%. Finally, through chip-based design and modularization, the three system-level problems of "lifetime advantage, readout power threshold, and replenishment throughput during operation" will be solved, and an extrapolable scale upper limit will be realized.
Regarding commercial implementation, Fang Xiaotian believes that to realize the real commercial value of the neutral-atom route, the "delivery" problem must be solved. The future quantum computer cannot be a "delicate device" that requires a team of doctors to maintain on site for weeks and can only be used for a few hours, but an engineering product that can be quickly deployed, operated and maintained with lightweight remote control, and connected to the cloud platform to provide external services.
In terms of scenario selection, Atomology currently prefers to combine with quantum computing centers (the "quantum-supercomputing integration" infrastructure), as well as fields with explosive demand for computing power such as biopharmaceuticals and financial optimization.
"Based on the long-term accumulated industrial experience of the team, our strategy is to carry out divergent verification and converge to focus, invest in several preferred directions, iterate engineering products through actual feedback, and finally converge to the core scenarios that generate real value." said Fang Xiaotian.
The following is an edited excerpt of the interview between Hard Krypton and Fang Xiaotian, COO of Atomology:
Hard Krypton: There are many companies working on neutral-atom quantum computing now. What is the core differentiated advantage of Atomology's route compared with its peers?
Fang Xiaotian: The biggest difference between us and most peers lies in the definition of "product". The approach of many teams now is to build quantum computers based on optical platforms, which is essentially making a more sophisticated scientific instrument. From the first principle, this is the difference in underlying logic caused by two different goals. The academic circle pursues the highest indicators under specific conditions, while the industrial circle requires stable and reproducible output even in the worst environment.
The underlying logic of Atomology is to combine atoms with photonic chips. Traditional optical tweezer arrays are limited by the physical limits of spatial light modulators such as AOD and SLM, and the upper limit of scale is usually at the order of 10,000, with frequency mixing and crosstalk. We use metasurface photonic chips to generate millions of static optical tweezers on a single chip at one time, and combine high-density optical waveguide chips and on-chip regulated photonic chips to achieve integration and stable operation while ensuring performance. At the same time, for peripheral devices that cannot be made chip-based, we have made highly standardized modules. This "chip-based + modular" system endows our system with industrial-grade attributes of assembly line manufacturing, rapid iteration, easy deployment and maintenance.
Hard Krypton: Quantum computing is generally considered to be far from large-scale commercial implementation. How does Atomology consider its commercialization pace and scenario selection?
Fang Xiaotian: Our requirement for ourselves is to focus entirely on the R&D of engineering products in the next 2 to 3 years, first develop absolute computing power and engineering stability, and create a deliverable complete machine that can be truly delivered. It doesn't make much sense to talk about short-term revenue now, but we are not worried about the recent marketization, because the team has accumulated a lot of technology in Shanghai Jiao Tong University and the industry, which lays a foundation for our subsequent commercialization.
In terms of application scenarios, we value two lines: one is to combine with quantum computing centers (such as "quantum-supercomputing integration" centers). This type of national or regional infrastructure has extremely high requirements for system scale and stability, which is very suitable for our ultra-large-scale arrays; the other is industries with explosive demand for computing power such as biopharmaceuticals and financial optimization. Our recent strategy is "divergent verification and convergent focus", investing in several preferred directions, iterating our engineering products through actual feedback, and finally converging to the core scenarios that generate real value.
Hard Krypton: What is the technology evolution direction of neutral-atom quantum computing in the next few years? What are the key milestones for Atomology next?
Fang Xiaotian: Neutral atoms will inevitably evolve in three directions: large-scale, reconfigurable and high-fidelity. This is also our core roadmap. In terms of scale, we first realize static arrays of 100,000 to 1,000,000 levels through metasurface chips; in terms of reconfigurability, we use on-chip modulators and 3D optical waveguides to achieve high-speed, high-precision independent control of light beams; in terms of fidelity, we continuously improve atomic lifetime and gate operation accuracy by combining low-temperature environment, continuous loading and ultra-high-speed low-noise imaging technology.
Our recent engineering goals are very clear: in the first phase (M1), we will achieve stable continuous replenishment and high-fidelity imaging closed-loop of 100,000-level arrays; in the second phase (M2), we will advance to million-level arrays and complete the architecture verification of partitioned processors (storage, interaction, readout, recycling). Quantum computing is a long-distance race. Our task at this stage is to take the most difficult but correct path of "engineering" steadily, so that China can have independent and controllable core hardware capabilities in the competition of next-generation computing infrastructure.
Investor's View:
China Internet Investment Fund said: "Quantum computing is a key field for China to seize the commanding height of future science and technology and achieve high-level self-reliance and self-improvement in science and technology. At present, continuous breakthroughs have been made in the global neutral-atom route. Atomology has neutral atom, photonic chip, metasurface and industrial engineering experience at the same time, and builds core capabilities around scale and engineering, which is also the direction required by the development of China's quantum computing industry. We look forward to the company adhering to independent innovation, steadily promoting technology verification and industrial exploration, and contributing to the innovative development of China's quantum computing industry."
The investment team of Muhua Tech Venture Capital said: "Quantum computing is at a critical stage from scientific research breakthrough to industrial breakthrough. The neutral-atom route has both scaling capability and high-fidelity potential, and is an important direction for general-purpose quantum computing in the future. The competition of quantum computing in the future is not only the competition of physical systems, but also the competition of engineering capabilities, industrial collaboration capabilities and ecological construction capabilities. Atomology has both profound technical accumulation and clear engineering route. It not only has long-term accumulation in the direction of neutral atoms, but also forms unique advantages in the fields of photonic chips, metasurfaces and system engineering. This complete capability system from underlying devices to system platforms has extremely high value in the future industrialization competition."