D-Wave Makes Gate-Model Power Move With Quantum Circuits Buy

In the early days of D-Wave’s history, the company made a decision to pursue annealing as its first technology to build a quantum computer because it promised to offer the fastest path to commercial quantum computing.

That’s what Trevor Lanting, the company’s chief development officer, reminded us this week as the company made a big move to expand into gate model quantum computing, extending the number of use cases it can address with its quantum machines.

“The last fifteen-plus years have really proven that out,” Lanting told The Next Platform this week. “But we see annealing as hitting a couple of core, important use cases, like optimization. Quantum optimization is a critical one. We’re just starting to see some new use cases around accelerating machine learning and potentially a quantum proof-of-work for a blockchain technology that uses our annealing technology.”

The plan worked. For several years, D-Wave has provided organizations commercial access to its Advantage annealing quantum technology through its Leap cloud-based platform. The workloads have included studying protein folding, directly mirroring the microscopic interactions of electrons, and developing a greater understanding of quantum physical processes during the formation of the universe. Almost a year ago, D-Wave took another significant step, selling a 5,000-qubit Advantage system to the Jülich Supercomputing Centre at Forschungszentrum Jülich (FZJ) in Germany.

As many other quantum companies are still deep in the research and development phase, D-Wave has a commercial business that in the third-quarter last year saw revenue double year-over-year, from $1.8 million in 2024 to $3.7 million in 2025. There also was a jump of 80 percent in closed bookings, which hit $2.4 million in Q3 after reaching $1.3 million the previous year.

That said, D-Wave executives continue to pursue their dual-platform strategy of providing not only annealing systems but also gate model quantum computers using the superconducting modality that others – such as IT giants IBM and Google as well as various pure-play vendors like Rigetti, IonQ, and SpinQ – are advancing.

D-Wave this week made two significant steps in advancing its gate model quantum ambition that chief executive officer Alan Baratz said will mean that the company will “allow us to bring a scaled, error-corrected, superconducting gate model quantum computer to market ahead of anybody else. That’s a very strong statement, but one that we firmly believe in.”

The company today said it is spending $550 million – $300 million in D-Wave common stock and $250 million in cash – to buy startup Quantum Circuits, a company that spun out of Yale University more than ten years ago that builds error-correcting superconducting qubits. Quantum Circuits uses what it calls a “dual-rail” technology that includes error correction built in that not only improves the quality of the qubits but means that fewer physical qubits will be needed to create logical qubits.

Error correction is a significant hurdle that needs to be cleared to reach fault-tolerant, scalable quantum computing. Qubits are fragile and can easily lose their quantum states, a process known as decoherence. They can be broken apart by a range of disturbances in their environments, from temperature and vibration to sound and stray particles. Any of these can cause errors during computing, which would make reliable quantum systems impossible.

“The idea behind error correction is that you can duplicate the number of physical qubits that are required to encode one bit of information, but exponentially suppress the errors as you add more and more of this redundancy,” Rob Schoelkopf, co-founder and chief scientist of Quantum Circuits, said during a media briefing announcing the deal.

The dual-rail architecture allows extra information to be added to the error correction. As logical qubits are built, the errors will be suppressed more quickly than what’s seen in other superconducting systems or other platforms, Schoelkopf said.

“The qubit is encoded by a single microwave photon that is shared between two superconducting cavities or resonators,” he said. “This gives us the usual abilities of quantum bits – a zero and a one and superpositions – but it also gives us a third state in which we can detect when the photon was lost. This means that this dual-rail qubit has error correction built in. It’s a unique functionality. … What this means is that we can first use the dual-rail qubits themselves as physical qubits, use the error detection to obtain … the kind of fidelities that are usually associated only with [trapped ion] trap quantum computers, but with the speed a thousand times faster and the scalability of superconducting platforms.”

D-Wave’s Baratz said Quantum Circuits’ dual-rail technology “is fundamentally changing how we can think about and pursue error correction in the sense that it will, on the one hand, allow us to use error detection as a part of the development of quantum algorithms, and on the other hand, it will allow us to error correct quantum systems with far fewer physical qubits per logical qubit.”

The CEO said D-Wave will fold Quantum Circuits into its business but added at the Quantum Circuits team and the work its does will remain at its site in New Haven, Connecticut, with D-Wave adding more people and improve the lab.

Schoelkopf, who developed both the dual-rail and transmon qubit technologies – a transmon superconducting qubit essentially stores quantum data in its lowest energy levels, 0 and 1 – said Quantum Circuits already has alpha users working with its 17-qubit dual-rail transmon system called “Seeker” and said the system will be generally available this year, along with a software toolkit for developing quantum algorithms.

Next year will come a 49-qubit dual-rail system with dual rail solvers on D-Wave’s Advantage platform. A 181-qubit dual-rail system – that will be designed to eventually expand to as many as 1,000 qubits – will come in 2028.

In its work developing a superconducting quantum computer, D-Wave has been spending more of its time developing advanced cryogenic packaging to help scale both superconducting and annealing systems than it has on error correction. The company unveiled a development effort around cryogenic packaging last July as a key part of its dual-platform strategy.

Advanced Cryogenic Controls

The day before announcing the deal to buy Quantum Circuits, D-Wave said it had process that allowed for on-chip cryogenic controls of qubits and multiplexed digital-to-analog converters (DACs) to reduce the number of bias wires needed to communicate with qubits – only 200 wires are needed control tens of thousands of qubits and couplers – bump bonding that stacks a quantum processor with a control chip.

The development will help drive greater scalability for quantum system, D-Wave’s Lanting said. One factor with superconducting quantum systems is that the qubits need to run in an extremely cold environment. The more wires needed to connect to and communicate with qubits, the more heat is created. With D-Wave’s process, few wires will be needed for each quantum processing unit (QPU).

Lanting said D-Wave’s annealing quantum systems already included processes to allow them to control tens of thousands of devices with only a couple of hundred control lines.

“We’ve taken all of this scalable control of qubits that we built for our annealing architecture and we’ve basically adapted it and are able to control our gate model qubits with the same technology,” he said. “For our annealing processors, DACs are built into the fabric of the processor. For our gate model systems, what we’ve done is taken a DAC chip and then used superconducting bump bond technology to interconnect it to a high performance qubit chip. Everything is at the cryogenic stage, but this local cryogenic control is really what allows you to reduce the number of lines that go into your cryogenic enclosure to control your devices.”

This will allow for the kind of scalability that will be needed as quantum systems grow to where they need to support a million or more qubits.

Many of the investments D-Wave is making in its innovation efforts will benefit both its annealing platform and its gate-model ambitions, Lanting said, noting the cryogenic packaging work. He estimated that 60 percent of the company’s patent portfolio covers both architectures.

“We have been able to leverage a significant amount of that investment for our gate-model architectures,” he said. “The fundamental technology is very similar, so it has to be adapted to be able to control gate-model devices.”

The Dual-Platform Advantage

The combination of annealing and gate-model quantum will give D-Wave an advantage over competitors, executives said. There are jobs that annealing systems are particularly good at, while others – such as quantum chemistry or high-performance material simulation – that a gate-model quantum computer will run better. However, both Baratz and Lanting said that as they look over a quantum landscape that includes a range of modalities, including trapped ions and neutral atoms, they believe that superconducting qubits will eventually emerge the gate-model of choice.

Superconducting has two key advantages, Lantin said. The first is speed, where superconducting is as much as 1,000 times faster than trapped ion or neutral atom approaches.

“The second reason is that we are basically harnessing the manufacturing base, the manufacturing techniques, and the packaging techniques that have been built up to produce our semiconducting world: the CMOS chips, the CPUs, the GPUs,” he said. “All of that fantastic explosion of capability is based on photolithic graphic processes and packaging processes that we can harness for our superconducting approaches.”