Quantum Computers May Soon Overtake Classical Systems
A new benchmark test reveals how close quantum processors are to outperforming classical computers, showcasing real-world potential.
Quantum Computers Near Milestone as New Benchmark Reveals Real-World Strength
A groundbreaking benchmarking study has brought the world one step closer to understanding when quantum computers will eclipse their classical counterparts. The test, which evaluated quantum processing units (QPUs) from five leading vendors, sheds light on which systems are advancing rapidly—and which still have ground to cover.
Measuring Progress in Quantum Hardware
The research, conducted by scientists at Germany’s Jülich Research Centre, analyzed 19 QPUs from IBM, Quantinuum, IonQ, Rigetti, and IQM. Their goal: to create a universal yardstick for evaluating QPU performance. The result is a test that assesses both width (number of qubits) and depth (complexity of quantum circuits involving two-qubit gates).
Two-qubit gates are critical operations in quantum systems, acting on entangled particles and forming the building blocks of real-world quantum applications. As circuits grow in depth and complexity, they begin to push the limits of what even today’s best classical computers can simulate—marking a potential threshold for quantum supremacy.
IBM and Quantinuum Lead in Distinct Areas
According to the results, IBM’s chips outperformed others in managing deeper circuits, a reflection of the company’s ongoing refinement of its QPU architecture. Notably, the Heron chip showed remarkable improvements over its predecessor, the Eagle. This was attributed to advancements in both hardware design and software integration—particularly the use of fractional gates that reduce circuit complexity.
Quantinuum, meanwhile, excelled in width. Its H2-1 chip successfully ran benchmarking algorithms on a staggering 56 qubits, a feat that makes classical simulation nearly impossible. This suggests that, in certain use cases, quantum machines may already be exceeding the capabilities of traditional supercomputers.
Real-World Algorithms in Action
The team used a variant of the Quantum Approximate Optimization Algorithm (QAOA) known as LR-QAOA to assess how well QPUs handle practical workloads. This algorithm was applied to the MaxCut problem, a notoriously difficult optimization challenge where nodes of a graph must be divided to maximize the edges between two groups.
Quantinuum’s H2-1 chip executed this with three layers of LR-QAOA—totaling over 4,600 two-qubit gates—while IBM’s Fez chip tackled even deeper circuits, involving nearly a million gates. Remarkably, Fez retained meaningful data up to 300 circuit layers, indicating significant depth tolerance.
Smaller Players Show Promise, but Challenges Remain
Not all QPUs tested met the benchmark’s expectations. Rigetti’s Ankaa-2 chip underperformed in both width and depth, signaling that while the field is evolving fast, some platforms may require more time to mature.
Still, the fact that companies beyond the industry giants are making measurable progress speaks to the growing competitiveness in quantum hardware development. These strides hint at a future where quantum computing becomes accessible across industries—from pharmaceuticals to finance to materials science.
An Evolving Standard for a Rapidly Evolving Field
The benchmark itself is part of the innovation. Built to be platform-agnostic, it uses a scalable, repeatable, and cost-effective approach. The researchers designed it to offer consistent metrics regardless of the QPU manufacturer, ensuring a fair comparison across technologies.
Yet, it’s not without limitations. For example, the test uses fixed parameter schedules, which can’t be optimized in real time—a constraint the researchers acknowledge. They call for more benchmarking models to be developed, each capturing different aspects of QPU performance.
The Road Ahead for Quantum Performance
This new benchmark offers more than just scores—it offers a vision of where quantum computing is heading. The performance of systems like IBM’s Heron and Quantinuum’s H2-1 suggests we may be approaching a tipping point. Soon, for some applications, quantum machines might not just compete with—but outperform—classical supercomputers.
As quantum hardware continues to mature, expect benchmarking to become the linchpin of innovation, guiding both investment and scientific breakthroughs. The tools may be evolving, but the mission remains the same: to unlock computing power beyond anything we’ve known.
Disclaimer:
This article is for informational purposes only and does not constitute financial, scientific, or technical advice. The benchmark results referenced are based on preliminary research and may evolve with future updates or peer review.