Picture Aiko Tanaka, a 29-year-old materials researcher at a small lab in Osaka, running the same molecular simulation for the third week straight because her university’s classical computers can’t handle the math fast enough. Across town this summer, that kind of wait time took a real step toward getting shorter. Japan switched on its first operational neutral-atom quantum computer, a machine researchers hope will eventually cut simulations like hers from weeks to days.
Researchers at the Institute for Molecular Science (IMS), led by Professor Kenji Ohmori, brought the new machine online this week. They named it Shunkai. Its quantum processing unit comes from Infleqtion, the only foreign quantum partner Japan’s Science and Technology Agency selected for its Quantum Moonshot program.
How a Neutral-Atom Quantum Computer Works
A neutral-atom quantum computer traps individual atoms in place with tightly focused lasers, then uses those atoms as qubits. The approach skips much of the bulky wiring that superconducting chips need. It also avoids some of the extreme cooling those chips require. Cooling and wiring limit how large a quantum machine can grow, and that ceiling costs labs real money in equipment and floor space. IBM hit this same wall earlier this year and answered it with redesigned cryogenic modules that let its chips scale without choking on cable count.
Infleqtion’s neutral-atom design takes a different route around the same problem. Lasers hold the atoms steady, and researchers reconfigure them without redesigning the whole chip. Infleqtion CTO Pranav Gokhale said the QPU “delivers the programmability, scalability and fidelity control that next-generation systems demand.” That combination of precise control and room to grow turns a lab demo into a platform other researchers can actually build on.
Scaling From 50 Qubits Toward Fault Tolerance
Shunkai started running with roughly 50 qubits. Infleqtion and IMS plan to push that number to around 500 in the near term. Their longer roadmap aims much higher: a fault-tolerant machine with up to 10,000 physical qubits, backed by real quantum error correction. Error correction remains the industry’s hardest problem, and different teams keep attacking it from different angles. A recent AI-powered decoder outpaced standard benchmarks on live quantum hardware data just last week.
What sets Shunkai apart is that it already runs, instead of sitting on paper as a proposal. Japan’s team plans to open external access to the machine. That will let outside researchers, including scientists working on problems like Aiko’s, develop applications and test error-correction techniques directly on real hardware, according to The Quantum Insider’s report on the launch. That access turns a single national lab’s project into a shared resource other scientists can build on, instead of a machine only a handful of insiders ever touch.
Why It Matters Beyond the Lab
Faster simulations do more than save researchers time at a keyboard. They can shorten the path to new drugs, better batteries, and cheaper fertilizer, all of which eventually show up in lower prices for everyday buyers. A university lab that once needed months of supercomputer time to model a new material could get comparable answers in days once quantum hardware like Shunkai matures. That is months of grant money and staff hours freed up for other work.
Japan is building its quantum industry largely from scratch. If Infleqtion and IMS hit their next milestone of roughly 500 qubits, Shunkai could become the machine other countries measure their own quantum progress against.