In a Caltech lab, a computer screen showed thousands of tiny points of light—each one a single atom, held in place by laser beams. This striking image revealed 6,100 stable quantum bits, or qubits. It’s the largest neutral-atom array ever created and a key achievement for quantum computing. The previous record for a neutral atom array was just 1,180 qubits.
“This is an exciting moment for neutral-atom quantum computing,” says Manuel Endres, the physicist who led the new study, published today in Nature. “We can now see a pathway to large error-corrected quantum computers. The building blocks are in place.”
Credit: Caltech/Endres Lab
A Big Leap for Tiny Particles
Quantum computers, unlike conventional machines, harness the weird rules of quantum physics. Their fundamental units are called qubits, unlike bits in “normal” computers. Qubits can exist in two states at once, a property called superposition. This makes them extraordinarily powerful for solving certain problems, but also incredibly fragile. Small disturbances like thermal noise, stray photons, even vibrations can knock a qubit out of its delicate quantum state.
To make practical quantum computers, researchers must increase the number of qubits, maintain their coherence, and control them with high precision. That’s what makes the Caltech achievement so remarkable.
Using optical tweezers (laser beams shaped to trap single atoms) the team split a single high-powered laser into 12,000 tweezers, each holding a cesium atom suspended in a vacuum. From those, they stably loaded 6,139 atoms into a compact grid only a millimeter across.
“On the screen, we can actually see each qubit as a pinpoint of light,” says Hannah Manetsch, one of the Caltech graduate students who led the study. “It’s a striking image of quantum hardware at a large scale”.
Quantity and Quality
Scaling up quantum systems often comes at the expense of quality. More atoms typically mean more noise, more complexity, and more instability. But here, the researchers managed to keep their qubits in superposition for an average of 13 seconds—nearly 10 times longer than in previous systems of this kind. They also manipulated individual atoms with 99.98% accuracy.
“Large scale, with more atoms, is often thought to come at the expense of accuracy,” says co-lead author Gyohei Nomura. “But our results show that we can do both. Qubits aren’t useful without quality. Now we have quantity and quality”.
The team also demonstrated that they could move these atoms hundreds of micrometers (about the width of a human hair) while keeping them in superposition. That mobility is crucial for a specific kind of architecture known as zone-based quantum computing. This approach could simplify how quantum circuits are built and how errors are corrected.
Manetsch likens the challenge to a balancing act. “Trying to hold an atom while moving is like trying not to let the glass of water tip over. Trying to also keep the atom in a state of superposition is like being careful not to run so fast that water splashes over”.
The team achieved a record-breaking coherence time of 12.6 seconds—the longest ever for hyperfine qubits in an optical tweezer array. They also maintained an imaging survival rate of 99.99%, meaning atoms remained trapped and readable for long durations. These are critical metrics in building any scalable quantum system.
Building Toward Entanglement
But this doesn’t solve all our quantum computing problems.
The next big step is entanglement—the phenomenon that Einstein famously called “spooky action at a distance.” Entangled qubits behave as one, no matter how far apart they are. It’s the secret sauce that gives quantum computers their true power: the ability to simulate natural systems governed by quantum mechanics, such as molecules, materials, or even space-time itself.
The researchers have laid out a clear plan to connect these 6,100 atoms into a coherent, entangled system capable of full-scale quantum computations. They estimate that by improving their laser systems, scaling up optical hardware, and employing smarter loading algorithms, they could increase their qubit count to over 10,000 in the near future—and perhaps a million within a decade.
For now, despite this achievement, quantum computers remain an elusive promise. Existing machines, like those developed by IBM, Google, and Atom Computing, have demonstrated limited quantum advantage, but with relatively few qubits and short coherence times.
“This is an amazing demonstration of the simple scaling that neutral atoms have to offer,” said Ben Bloom of Atom Computing, who was not involved in the study, as per New Scientist.
Mark Saffman, a quantum physicist at the University of Wisconsin–Madison, called the result “encouraging” but noted that more experimental work is needed before this becomes a full-fledged quantum computer.
Still, this work pushes the boundary. It offers a realistic glimpse at how quantum hardware could evolve—not in decades, but in years.
A Window Into the Quantum World
In addition to building quantum computers, these efforts also help us better understand the physics behind them. For Manetsch and her colleagues, the thrill is about exploring new frontiers in physics.
“It’s exciting that we are creating machines to help us learn about the universe in ways that only quantum mechanics can teach us,” she says.
On the screen, each pinpoint of light marks a single atom in quantum motion. For the Caltech team, the array is more than a prototype computer—it’s a tool for exploring the fundamentals of nature.
And now, they have 6,100 of them.
