What’s next on our roadmap and why it includes gate-model quantum computing
D-Wave has never been one to choose the easy path when it comes to our quantum roadmap. Fifteen years ago, we chose the path less traveled (and less believed in) when we decided to pursue quantum annealing. We made that choice for one primary reason: we believe annealing is the fastest approach to deliver practically valuable applications for business customers. Why? When it comes to the quantum computing technology landscape, quantum annealing is uniquely designed for optimization problems, and this problem class makes up a very significant proportion of the enterprise problem space.
Our mission of practical quantum computing has evolved even further into a hyper-focus on being #relentlesslyrealistic in whatever we do, and this worldview informs every design, development, and business decision we make.
Today’s product expansion
Today is a big milestone for us for a few reasons, one of them being the rollout of key new products and enhancements. We announced the general availability of a robust performance update to Advantage™, our flagship quantum computing system built for business. Just a year after the launch of Advantage, we’ve continued to expand its value via a newly fabricated and performant QPU. Available today in the Leap™ quantum cloud service, the new system update includes several advancements that help customers solve larger and more complex problems faster and with greater precision.
Additionally, we’ve unveiled an all-new hybrid solver — the constrained quadratic model (CQM) — which expands our hybrid solver services in Leap and shows just how easy it is to get started building in-production quantum hybrid applications today. The CQM solver directly incorporates problem constraints in the natural way the problem is expressed, rather than requiring additional steps to translate or transform it. Now, not only will users benefit from a simplified and more native expression of constrained problems, but they can also address larger problems overall.
Clarity, our technology roadmap for the future
But we’re also equally passionate about building for the future as we are delivering new technologies today.
So today, as we also release a preview of our new roadmap (code-named Clarity), our guiding principle of practical value has not changed. Clarity includes a myriad of enhancements and performance initiatives across our annealing Advantage system, Leap quantum cloud service, hybrid solvers, and perhaps most notable to those who have followed us over the decades, a product expansion into error-corrected, gate-model systems.
Delivering on Clarity will make us the first in the industry to offer both annealing and gate-model quantum computers via an integrated, full-stack quantum cloud platform. The expanded stack will have all of the cross-platform developer tools, templates, and resources built on top of it to enable users to leverage the best quantum approach for their particular use case. Why does this matter? We’ve learned that every technology decision must be informed by the intersection of the hardware and the software necessary to build, run and scale today’s (and tomorrow’s) powerful quantum systems. But being first isn’t our motivation. As it was in 2004, we’re doing this because it’s in line with our core mission to build the most practically valuable, and usable, quantum systems possible. Cross-platform, platform-agnostic offerings are what our customers are asking for and where we see the market heading.
Cross-platform tools and even more powerful solvers
We envision a day where customers run their problems in the cloud and benefit from the best of both classical and quantum via powerful, cross-platform solutions. Clarity is the path to get there. The roadmap includes enhancements and lower-noise versions of Advantage (the forthcoming Advantage 2 annealing system), more powerful and expanded hybrid solvers, cross-platform, open-source developer tools, and an error-corrected gate model system.
The initiative around cross-platform open-source developer tools will enable customers to invest in one tool’s platform and use it across multiple quantum systems. Further, more powerful hybrid solvers will fuel an array of new use cases while bringing the best of quantum and classical resources together for rapid application development in the cloud. Our quantum cloud service gets more powerful everyday with product enhancements informed by customer feedback. As more and more customers build in-production applications, it will continue to enable cross-platform use cases and applications at global scale. For example, imagine a pharmaceutical company utilizing quantum across their entire product lifecycle: gate-model systems will assist with drug discovery, while annealing systems will ensure patient trial optimization and enable faster routes to market in manufacturing and distribution of that new drug.
The business case: platform-agnostic offerings are what the market demands.
We’ve long argued the future of quantum computing isn’t about a single approach or theory. To us, it’s never been about annealing vs. gate-model or even quantum vs. classical. It’s about using the right solutions for the right applications to drive real-world value and business impact for our customers.
When we started, we saw that annealing would be the most direct path to practically valuable quantum applications. That prediction proved true. Today, customers have built more than 250 early applications using our systems for real-world problems ranging from auto manufacturing to protein design, financial portfolio optimization, and more. Annealing remains core to our roadmap for this reason.
We’ll continue to invest in and develop even more powerful annealing systems. We’ll do this because annealing quantum computers are uniquely designed for what looks to account for more than 30% of the overall quantum application market: optimization use cases. In fact, recent publications point to the fact that annealing is better for optimization both today and likely in the future. The pre-processing overhead and lesser performance of current gate-model systems make them ineffective for optimization. [See: “Applying quantum algorithms to constraint satisfaction problems,”¹ “Noise-Induced Barren Plateaus in Variational Quantum Algorithms,”²and “Training variational quantum algorithms is NP-hard — even for logarithmically many qubits and free fermionic systems”³).]
And like we did when we initially chose to pursue annealing, we’re looking ahead. We’re anticipating what our customers need to drive practical business value, and we know error-corrected gate-model quantum systems with practical application value will be required for another important part of the quantum application market: simulating quantum systems. This is an application that’s particularly useful in fields like materials science and pharmaceutical research.
Ultimately, as market demand for quantum computing continues to accelerate, the benchmark for investment is quickly becoming practical business value. Forward-thinking organizations don’t want to navigate the either/or of “VHS vs beta” in quantum computing. They want results, performance, and ROI. We hear you. And this, along with nearly 20 years of trials, errors, mistakes, and incredible successes, is where we started as we imagined Clarity.
We’ve always made roadmap and product decisions based on customer input and feedback. Designing a roadmap that meets the market need for the first platform-agnostic system is no different. Customers expect their quantum investments to make sound business sense today, while also ensuring that their quantum investments (people, resources, training, technology) have an enduring business benefit and pay dividends in the long term.
The technical case: designing a roadmap to finally solve for quantum’s scaling problem.
For these reasons, we don’t take the decision to pursue gate-model technology lightly. We chose not to pursue it years ago in large part because neither the theory nor the superconducting circuit technology were truly up to the task at the time.
Why do we think we can add it to our platform now? In the intervening years, two important things have happened. First, the theory has matured, and secondly, the team at D-Wave has worked to drive integrated superconducting circuit technology to new heights. Our scientific and engineering teams have worked tirelessly and creatively to solve some of the trickiest engineering challenges (our Advantage system’s 15-way connectivity is a testament to that), and our materials science program has provided us with new, critical insights that challenge previously held beliefs about microscopic sources of noise in the fabrication stack.
Now is exactly the right time from a technical and theory perspective to face the challenges of gate-model implementation head-on. We’re not doing it because it’s easy. We’re doing it because it’s necessary to the future of quantum computing and because we have the engineering and operational expertise to do it for our customers.
Looking ahead: balancing (cautious) optimism with measured deliberation.
The path forward, while not straight, will require us to lean on our team’s hard-won reputation as expert superconducting circuit engineers and our business experience delivering real-world solutions. This won’t be an overnight journey, but it’s a necessary one for the advancement of practical quantum computing. Along the way, we’ll be working closely with and listening to our customers to build out next generation systems, cross-platform tools, and powerful new hybrid solvers to accelerate growth across the quantum ecosystem. The pursuit of technological advancement is not easy, but our customers have always been the source of truth when it comes to what will be most valuable in the real world, and our team has a track record of turning those innovations into reality. The road ahead will continue to demand determination and clarity; our team is more than ready to work towards delivering on quantum computing’s full potential to customers.
Join us at D-Wave’s annual Qubits event October 5–7 to learn more about our roadmap and how businesses, developers, and researchers can harness the power of quantum computing today and in the future.
 E. Campbell, A. Khurana, A. Montenaro, “Applying quantum algorithms to constraint satisfaction problems”, Quantum 3, 167 (2019).
 Wang et al., “Noise induced barren plateaus in variational quantum algorithms” (2021)
 M. Bittel and L. Kliesch, “Training variational quantum algorithms is NP-hard — — even for logarithmically many qubits and free fermionic systems” (2021)