The News: In recent weeks, quantum computing companies Alice & Bob, Diraq, and Quantinuum have each made announcements that advance the state of the art in qubits in different ways. Alice & Bob touted its paper in Nature, demonstrating significantly improved times between bit-flip errors. Diraq announced 99.9% 1-qubit silicon quantum dot control accuracy using standard CMOS materials in a semiconductor foundry. Quantinuum advanced its flagship H2 system from 32 to 56 qubits, passing the point where gate-and-circuit simulation via state vectors is feasible. See the Alice & Bob, Diraq, and Quantinuum press releases for more details.
Quantum QuickTake: Qubit News from Alice & Bob, Diraq, and Quantinuum
Analyst Take: Across the many quantum bit (qubit) modalities, vendors have steadily improved the number and quality of their qubits. Although quantum computers today are small compared to what we will eventually need to achieve Practical Quantum Advantage, improving qubit quality by reducing error rates is mandatory. Furthermore, if vendors cannot ultimately manufacture quantum computing systems at scale, we will be left with individual hero machines. Finally, while increasing the number of qubits in a single quantum processing unit (QPU) is essential, work must begin to connect these QPUs before we litter the future world with tiny, useless quantum systems.
Alice & Bob: Cat Qubits and Bit Flips
A “cat qubit” is a design variation on a superconducting qubit that reduces bit-flip errors in single qubits. Recall that if we represent the superposition state of a qubit as
a |0⟩ + b |1⟩,
then two primary errors can occur. A bit flip interchanges a and b:
b |0⟩ + a |1⟩.
A phase flip replaces b with –b:
a |0⟩ – b |1⟩.
Classically, a bit flip switches 0 and 1 in a bit, which is the only kind of bit error that can occur. Therefore, a phase flip does not have a corresponding bit error, but if we think more generally, we can imagine it as replacing a real number by –1 its value.
Both kinds of errors can happen when environmental noise, imprecise control, or manufacturing defects change the quantum state. Several vendors and academic groups are investigating cat qubits, including Alice & Bob and Amazon.
Ideally, we would reduce the occurrences of both error types to zero or very close to it. Alternatively, can we exponentially reduce one type, allowing us to focus on the other? In its Nature paper, Alice & Bob stated that they had increased the time between bit-flip errors to more than 10 seconds. This is an extraordinarily long time for superconducting qubits, far better than seen in non-cat systems from vendors such as Google, IBM, and Rigetti Computing.
Is there a gotcha? Alice & Bob still must reduce the phase flip error rate significantly. Moreover, their design must advance to having high-fidelity 2-qubit operations such as CNOTs. The bit flip improvements are essential but only part of the story.
Think of it this way: Suppose we need three factors, A, B, and C, to have excellent stats. We improve A to be extraordinary while leaving B and C the same or making them worse. Follow-up work can focus on B and C.
Diraq: Silicon Quantum Dot Qubits and Manufacturability
Diraq, a quantum computing company based in Sydney, Australia, works in the area of silicon quantum dots. Just as today’s classical semiconductors represented by memory, CPUs, and GPUs are outstanding examples of implementing billions of transistors in a single chip, the hope is that we can someday fit thousands, millions, or even billions of qubits on a single chip.
That’s the promise or hope, anyway. Like almost everything with quantum computing, theory is important, but working implementations are the proof points. Still, it’s an exciting potential way of scaling systems, and I am bullish on it. Progress on quantum dots has been slower than some other modalities, but the winners will not be decided in the next few years but rather in the next decades.
Diraq announced it achieved “a record control accuracy of 99.9% for a quantum bit (qubit) manufactured by imec using industry-standard CMOS materials on a 300mm silicon wafer.”
Regarding the news, I posted on LinkedIn:
“Aside from the strong 1-qubit fidelities, this Diraq announcement highlights an important dimension for #quantumcomputing: manufacturability. While there is only so much one can do with small quantum computers today, developers of any qubit modality must prove that they can go from academic theory to prototype to first commercial MVP to the manufacture of multiple systems with a large number of high-quality, well-connected qubits.”
If you are a quantum hardware startup and plan to someday go from Series B to Series C funding, you will likely have to demonstrate two things:
- You are on a short path to tens of millions of dollars of revenue.
- You have the manufacturing capability and supply chain to bring sufficient systems to market to meet demand and your revenue targets.
Diraq is very smart in ensuring that standard foundry processes can manufacture their devices. They can continue their innovation while maintaining their manufacturing potential.
In a comment exchange on LinkedIn, Andre Saraiva of Diraq assured me that while this announcement focused on 1-qubit control accuracy, their lab fidelities for 2-qubit operations are very good. I look forward to hearing more about those and how they are increasing the number of qubits.
Quantinuum: Trapped Ion Qubits and Scaling
When I first got involved with quantum computing in 2016, 50 was the magic number for qubits. Many thought that once a quantum computer exceeded this value with high-quality qubits, quantum computers could demonstrate capabilities impossible for classical computers. This led to the so-called “quantum supremacy” experiments and a much better understanding of the necessary utility of what quantum computers do versus classical systems.
Why 50? The most straightforward way of simulating quantum circuits classically is via state vectors (see my Dancing with Qubits, Second Edition book, section 11.8). The quantum state for one qubit requires two complex numbers. Every time we add a qubit, we double the number of values necessary. So, two qubits need 4 complex numbers, three qubits need 8, and ten qubits need 1024.
For 50 qubits, we need 1,125,899,906,842,624 complex numbers. This is one quadrillion one hundred twenty-five trillion eight hundred ninety-nine billion nine hundred six million eight hundred forty-two thousand six hundred twenty-four. In Python, a complex number uses 32 bytes, so our single quantum state using all 50 qubits would require 32 petabytes. Now, I know there are more efficient storage and simulation methods, but using the simple state vector representation is untenable classically.
For this reason, when any vendor passes this threshold, it feels it has moved into significant quantum territory, and this is what Quantinuum has done in moving from 32 qubits to 56. It announced
“With this upgrade in our qubit count to 56, we will no longer be offering a commercial ‘fully encompassing’ emulator – a mathematically exact simulation of our H2-1 quantum processor is now impossible, as it would take up the entire memory of the world’s best supercomputers.”
As part of its press release, Quantinuum gave an estimate for the reduction in power consumption for a quantum calculation compared to its classical counterpart. This is now required for me: don’t just tell me about qubits, circuit depth, and fidelities; talk to me about sustainability and power use.
Modularity Must Be Next
We have seen above various essential improvements involving qubits’ quality, number, and manufacturability. Here is my challenge to quantum system vendors: once you get the number of high-fidelity, well-connected qubits over 50, start putting an equal amount of work into modularity. That is, build multiple QPUs and connect them. Build larger quantum systems from several QPUs.
While providers of some modalities, including neutral atoms, quantum dots, and topological qubits, believe they can ultimately put thousands of qubits in a single QPU, that is a long way off. Start now with modularity. Several vendors are doing this now. They should bring more public attention to their efforts.
Key Takeaway
Mitigating errors, increasing the number of qubits, and ensuring the manufacturability of quantum computing systems are worthwhile and notable innovations. Modularity is becoming just as important, and vendor efforts need to move quickly in that direction.
Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold an equity position in any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.
Other Insights from The Futurum Group:
Quantum in Context: A Qubit Primer
Quantum in Context: Quantinuum and the Quest for More 9s
Quantum in Context: Microsoft & Quantinuum Create Real Logical Qubits
Author Information
Dr. Bob Sutor has been a technical leader and executive in the IT industry for over 40 years. Bob’s industry role is to advance quantum and AI technologies by building strong business, partner, technical, and educational ecosystems. The singular goal is to evolve quantum and AI to help solve some of the critical computational problems facing society today. Bob is widely quoted in the press, delivers conference keynotes, and works with industry analysts and investors to accelerate understanding and adoption of quantum technologies. Bob is a Consulting Analyst at The Futurum Group. He helps clients understand sophisticated technologies in order to make the best use of them for success in their organizations and industries. He is also an Adjunct Professor in the Department of Computer Science and Engineering at the University at Buffalo, New York, USA. More than two decades of Bob’s career were spent in IBM Research in New York. During his time there, he worked on or led efforts in symbolic mathematical computation, optimization, AI, blockchain, and quantum computing. He was also an executive on the software side of the IBM business in areas including middleware, software on Linux, mobile, open source, and emerging industry standards. He was the Vice President of Corporate Development and, later, Chief Quantum Advocate, at Infleqtion, a quantum computing and quantum sensing company based in Boulder, Colorado USA. Bob is a theoretical mathematician by training, has a Ph.D. from Princeton University, and an undergraduate degree from Harvard College.
He’s the author of a book about quantum computing called Dancing with Qubits, which was published in 2019, with the Second Edition released in March 2024. He is also the author of the 2021 book Dancing with Python, an introduction to Python coding for classical and quantum computing. Areas in which he’s worked: quantum computing, AI, blockchain, mathematics and mathematical software, Linux, open source, standards management, product management and marketing, computer algebra, and web standards.