Four qubits, entangled in just 14.8 microseconds. That is a few thousandths of the time it takes to blink. And the qubits sat in a diamond at room temperature, with no liquid helium and no dilution refrigerator. Most superconducting quantum computers need cooling near absolute zero to work at all. This diamond system did not. It is a small lab demonstration, but it hints at who might get to use some quantum technology next.
What did the Penn team actually do?
Researchers at the University of Pennsylvania made the advance, and the paper appeared in Nature Nanotechnology. The authors are Joseph Minnella, Mathieu Ouellet, Amelia Klein, and Lee Bassett. They built on nitrogen-vacancy centers, tiny defects in diamond that trap an electron and can act as a qubit. Three nearby carbon-13 atoms each carry a faint nuclear spin and act as extra qubits.
Scientists usually link qubits like these one pair at a time. It works, but it is slow, and every extra step gives errors room to creep in. Bassett’s lab tried something different. Instead of fighting the crosstalk between nearby spins, they used it. A single carefully timed pulse sequence entangled all four qubits at once.
How much faster is it?
The authors report 14.8 microseconds for a four-qubit entangled state, about 10 times faster than sequential two-qubit gates. Fidelity rose from 0.69 to 0.92. The new method was both faster and more accurate here, which is unusual, since speed and accuracy often trade off in quantum research. The authors say the gate runs close to the limit set by the electron-nuclear interaction strength.
| Method | Time to entangle 4 qubits | Fidelity |
|---|---|---|
| Sequential (old way) | About 139.9 microseconds | 0.69 (plus or minus 0.03) |
| Parallel (new way) | 14.8 microseconds | 0.92 (plus or minus 0.04) |
Source: The Quantum Insider and the authors’ arXiv paper. The sequential time comes from The Quantum Insider.
“About 10 times faster” is rounded. Using The Quantum Insider’s numbers, 139.9 divided by 14.8 is about 9.5. Outlets also differ on the baseline: The Quantum Insider lists the sequential time as about 139.9 microseconds, while Quantum Computing Report lists roughly 145 microseconds. They differ on three-qubit fidelities too.
Are these qubits really working at room temperature?
Yes. The system uses nitrogen-vacancy centers in diamond and runs at room temperature, with no cryogenic cooling. Superconducting quantum chips, like the ones IBM and Google build, need dilution refrigerators that cool them far below room temperature. That equipment is costly to buy and run. We found no sourced price in the Penn coverage, so we won’t quote one. The point is simple: a chip that works at room temperature skips that equipment.
IBM’s Nighthawk chip, which IBM plans to bring to ETH Zurich, is a superconducting design and still needs that cooling. Penn’s approach does not. The trade-off is that readout is less efficient and coherence is shorter at room temperature. Measurement errors are still worse than in a cooled lab, and cooling the diamond could improve measurement but would remove the room-temperature advantage. Fault-tolerant operation at room temperature is a future goal, not a result.
Why It Matters to You
You can’t buy or use this. It is four qubits in a single diamond, built for a research demonstration, not a commercial system, so don’t expect a diamond quantum laptop next year. The sensing side of this technology sits closer to real use, since NV-center diamonds already work as magnetic sensors. Take an illustrative case: a water utility technician who wants to find a cracked pipe without digging up the street. A cheap room-temperature sensor might one day help, but the Penn work does not demonstrate that. It shows a better gate in a small lab device.
What could this technology be used for?
The Quantum Insider lists these possible uses for faster multi-qubit gates. They are possible uses, not demonstrated products.
- Nanoscale NMR, a shrunk-down cousin of lab machines that analyze proteins and drugs
- Scanning magnetometry, for finding hidden defects in materials
- Quantum error correction, where faster gates could shorten circuits
Related quantum sensing work includes navigation tests that skip GPS. The Quantum Insider reported that the approach may extend to other defects in diamond, silicon carbide, or silicon. That has not been demonstrated. Quantum Computing Report adds that simulations of 500 sampled registers found parallel gates in most configurations.
What’s the catch?
This is a four-qubit lab demonstration. It did not show a large processor, a full error-correction cycle, or a useful computation. Scaling depends on finding or engineering nuclear-spin arrangements with the right coupling, and a five-qubit gate was not demonstrated. The result also depends on carefully tuned pulse timing, so it may not transfer directly to other setups. Watch what happens next: does anyone push this past four qubits? That will show how close room-temperature quantum tools are to practical use.
Has the work been peer reviewed?
Both outlets say it was published in Nature Nanotechnology. We read the arXiv version and the news coverage, not the journal paper itself.
Sources
- The Quantum Insider: Penn Researchers Demonstrate Room-Temperature Diamond System That Entangles Four Qubits 10 Times Faster (Sep 14, 2026)
- arXiv: Single-gate, multipartite entanglement on a room-temperature quantum register (Aug 11, 2025)
- Quantum Computing Report: University of Pennsylvania Demonstrates Single-Gate Parallel Entanglement on Room-Temperature Diamond Quantum Register (Sep 19, 2026)
