D-Wave Solves the 'Wiring Problem': Quantum Computers Get More Compact
D-Wave engineers have achieved a breakthrough in qubit control by placing cryogenic control directly on the chip, something previously considered science fiction. Now, instead of colossal cabinets of wires for large quantum systems, truly scalable machines can be built without sacrificing their key property—fragile precision.
A room-sized refrigerator? D-Wave just solved the problem that has choked quantum computers for decades.
Imagine a processor with a cable as thick as a python—and you need thousands of such cables. That's exactly what the architecture of gate-model quantum computers looked like until yesterday. On January 6, 2026, D-Wave announced a breakthrough that changes the game: the company demonstrated, for the first time in history, scalable cryogenic qubit control directly on the chip, without degradation of precision.
The solution came from an unexpected source—a technology that D-Wave had been refining for years on its adiabatic processors. In those systems, it controls tens of thousands of qubits and their connections through a mere 200 bias wires, thanks to multiplexed digital-to-analog converters. Now the same trick has been applied to gate-model qubits—and it worked.
"Without on-chip control and multiplexing, useful gate-model quantum computers require an impractically large number of wires and massive cryogenic enclosures," explained Dr. Trevor Lanting, Chief Development Officer at D-Wave. "Scalability is fundamental to the growth and increasing adoption of this technology. Controlling more qubits with fewer wires means building larger processors with a smaller footprint."
This is not a lab trick or a simulation. D-Wave used superconducting bump bonding and advanced cryogenic packaging techniques to build a multi-chip package combining a high-coherence fluxonium qubit chip with a multi-layer control chip. Key components were fabricated at NASA's Jet Propulsion Laboratory.
Why the 'Wiring Problem' Is a Problem at All
To understand the scale, a short excursion into the engineering hell that quantum hardware developers live in is needed. Qubits operate at temperatures near absolute zero—we're talking millikelvins inside dilution refrigerators. Meanwhile, control electronics live at room temperature. Each qubit requires an individual control line that runs from the warm world into the cold through multiple thermal shields.
The more qubits you add, the thicker this cable bundle becomes. At some point, it starts conducting heat into the cryostat faster than the cooling system can remove it. This is called "thermal load," and it sets a physical ceiling on scaling. You can't just add more wires—they'll burn your cryogenics.
D-Wave's solution is elegant in its simplicity: place digital-to-analog converters inside the refrigerator, next to the qubits, and communicate with them through multiplexed lines. One channel controls a group of qubits, not just one. The number of physical wires drops by orders of magnitude.
This is the moment when a lab curiosity turns into a product. "We believe this historic milestone positions D-Wave as the first provider of a truly scalable commercial gate-model system," added Lanting.
Two Technologies, One Company—and a Race for the Future
D-Wave has long been an outsider in the eyes of the mainstream quantum community. While IBM, Google, and IonQ chased universal gate-model processors, the Canadians sold adiabatic systems that excel at optimization problems but don't claim universality. They were called "not quite quantum," and sometimes harsher things.
The January 2026 breakthrough flips this perspective 180 degrees. It turns out that D-Wave's two decades of experience in superconducting quantum hardware—over 60% of the company's patents cover both architectures—provides a unique advantage precisely in the engineering challenges of scaling.
Three weeks after the announcement, at the Qubits 2026 conference, the company solidified its success. Usage of Advantage2 systems surged 314% year-over-year, and the Stride hybrid solver grew 114% in six months. And crucially, D-Wave confirmed plans to bring an initial gate-model system to market in 2026.
Added to this was the January acquisition of Quantum Circuits—a startup that brought dual-rail qubits with error detection to the portfolio. Such qubits can detect approximately 90% of errors at the moment they occur, with an erasure rate of 0.5%. CEO Alan Baratz, during the May earnings call, estimated that the technology could reduce the number of physical qubits per logical qubit to an order of magnitude.
Battle Map: Who's Winning, Who's Nervous
The balance of power after January looks like this.
IBM continues to lead in the number of deployed gate-model processors and the Qiskit ecosystem. But its approach requires classical cryogenic wiring, and the scaling problem is just as acute for it. Any solution that reduces wiring complexity is a potential threat to IBM's architectural foundation.
Google Quantum AI is silent about the details of its cryogenic architecture, but the race for quantum supremacy runs into the same physical limitations. The NASA JPL labs, where D-Wave fabricated its demonstration chip, are former partners of Google on the Sycamore project.
IonQ and Quantinuum work with trapped ions—and D-Wave delivers a direct blow to them. In the press release, the company specifically noted that superconducting qubits execute gates "significantly faster than trapped ions, neutral atoms, or photonics." The gap, they estimate, will become decisive as systems grow and precision improves.
On the financial front, the picture is mixed. First-quarter 2026 revenue plummeted 81%—from $15 million a year earlier to $2.86 million. But bookings soared 1994% to $33.4 million, including a $20 million system sale to Florida Atlantic University and the industry's first corporate QCaaS agreement worth $10 million. Baratz stated that the company now expects to sell two to three systems per year instead of one.
This is a classic scenario for a young tech company: revenue is volatile, bookings outpace revenue recognition, and R&D costs grow faster than revenue. Meanwhile, D-Wave has $588.4 million in cash on hand—93% more than a year ago.
What's Next: Roadmap to 2032
The May earnings call revealed specific numbers. D-Wave is targeting 175 physical qubits by the end of 2028, 10 logical qubits by 2030, and 100 logical qubits by the end of 2032. Demonstrated fidelities exceed 99.9% on a small system.
An initial gate-model system will appear in 2026—the company confirmed this twice, in January and May. Details are still scarce: no exact specifications, no price, no name. But Baratz mentioned that several customers have already expressed interest, including those wanting to buy a system and those seeking access via the Leap cloud service.
The most intriguing part is what will happen when D-Wave's gate-model system meets its own adiabatic machines in the same data center. The company already sells hybrid solvers that combine classical computing with quantum annealing. Adding a gate-model processor will create a three-layer architecture: classical, annealing, gate-model—with a common Ocean SDK and a unified cloud platform.
If D-Wave truly manages to be the first to deliver a commercially viable gate-model system with on-chip control, the definition of a "practical quantum computer" will have to be rewritten. And yes—that giant refrigerator stuffed with wires may soon become a museum exhibit.
— Editorial Team
No comments yet.