Researchers at the Wrocław University of Science and Technology are challenging claims of a quantum advantage in solving complex optimization problems. Their work demonstrates that a purely classical approach, the simulated bifurcation machine (SBM), achieves comparable or even superior scaling to quantum annealing when tackling quadratic unconstrained binary optimization problems. This nonlinear dynamical system, leveraging chaotic behavior instead of thermal fluctuations, shares key operational features with quantum annealing, including nearly parallel evolution and a defined relationship between energy gap, run-time, and solution quality. The team further found that small instances studied previously are insufficient to infer asymptotic behavior; extending the analysis to larger problems revealed robust classical performance, suggesting current quantum annealers likely won’t demonstrate a clear scaling advantage on these types of problems under the run-time accounting studied here. Quantumz.io Sp. z o.o. researcher J. Gardas of the Institute of Theoretical Physics, Faculty of Fundamental Problems of Technology, Wrocław University of Science and Technology, led the study, recently published in Phys. Rev. Applied volume 26, 014024, published July 8, 2026.
A classically-inspired computational model is challenging the anticipated supremacy of quantum annealers in solving complex optimization problems. This is notable because the SBM operates as a purely classical system, mimicking quantum processes without relying on superposition or entanglement. Quantumz.io Sp. z o.o. is affiliated with one of the researchers. Previous assessments of quantum annealing’s potential were based on analyses using small instances, a limitation the researchers addressed by extending their investigation to significantly larger problem sizes. However, the research isn’t a complete dismissal of quantum optimization; the team identified areas where future quantum devices could potentially outperform classical methods, but only if substantial reductions in hardware overheads are achieved. This suggests a narrow pathway for quantum optimization, contingent on overcoming current technological limitations and focusing on specific problem structures.
Recent analyses of quantum annealing’s potential to outperform classical algorithms in solving complex optimization problems are facing renewed scrutiny, as research demonstrates a classical system achieving comparable, and in some cases superior, results. A critical finding detailed in Phys. Rev. Applied is the inadequacy of previous studies for accurately predicting long-term performance. The research doesn’t entirely dismiss the potential for quantum optimization; sparse problem classes are identified as areas where future quantum devices could achieve an advantage, provided significant mitigation of hardware overheads occurs, representing a narrow pathway forward for quantum computing in this domain.
