IBM Says the Future Lies in Quantum-Centric Supercomputing

IBM has released a reference architecture for quantum-centric supercomputing that integrates quantum processors with classical computing systems. The blueprint outlines how quantum processing units (QPUs) can operate alongside CPUs and GPUs across research centres, on-premises infrastructure, and cloud environments.

The approach explains how coordinated workflows and open software frameworks can combine quantum and classical computing to address scientific challenges. It enables workloads that require both quantum algorithms and traditional high-performance computing resources.

These include problems in chemistry, materials science, and molecular simulation that exceed the limits of classical computing.

IBM said the system uses open software frameworks, including Qiskit, to coordinate workflows between quantum and classical systems. Developers and scientists can access quantum capabilities through familiar tools and computing environments.

Jay Gambetta, Director of IBM Research and IBM Fellow, said the approach builds on earlier ideas about quantum simulation. “More than four decades ago, Richard Feynman envisioned computers that could simulate quantum physics,” he said. 

“The future lies in quantum-centric supercomputing, where quantum processors work together with classical high-performance computing to solve problems that were previously out of reach.”

Researchers have already used IBM’s architecture in several scientific experiments. A team including IBM and the University of Manchester created a half-Möbius molecule and verified its electronic structure using a quantum-centric supercomputer.

The Cleveland Clinic simulated a 303-atom tryptophan-cage mini-protein, one of the largest molecular models run using this approach. Another project involving IBM, RIKEN, and the University of Chicago identified the lowest-energy state of engineered quantum systems.

IBM said its partner ecosystem will continue to refine the architecture. New algorithms and improved scheduling tools are expected to expand applications in chemistry, materials science, and optimisation.