IBM, Cleveland Clinic and RIKEN reach Gordon Bell finals with quantum protein simulation

Cleveland Clinic, RIKEN and IBM say their joint quantum computing research has moved into the finals for the 2026 ACM Gordon Bell Prize after modeling protein ligand systems at a scale that is still unusual for quantum hardware. The announcement matters less as an award update and more as a signal about where quantum assisted life sciences work is becoming concrete enough for serious high performance computing scrutiny.

The team says it modeled a 12,635 atom protein system, which IBM describes as the largest biologically meaningful molecule simulated using quantum computers. The work combined IBM Quantum Heron processors with classical supercomputers, including Fugaku at RIKEN and Miyabi G in Japan. The accompanying arXiv paper says the workflow used two 156 qubit quantum processors, ran with up to 94 qubits, executed 21,006 circuits over more than 239 hours, and collected 3.0 billion measurement outcomes.

Why this is useful beyond the prize list

Drug discovery depends on understanding how molecules move and how strongly they bind to targets. Those calculations are expensive because the electronic behavior of molecules follows quantum mechanics. The practical CyberOGZ read is that this is not a near term replacement for wet lab testing or classical simulation. It is a proof point for hybrid workflows where quantum processors handle selected chemistry subproblems while conventional supercomputers assemble and validate the larger picture.

The September revision to the paper reports two protein ligand complexes of 11,608 and 12,635 atoms. It also claims a system size increase of more than 40 times and accuracy gains of up to 210 times over the previous state of the art for this type of heterogeneous quantum and classical chemistry workflow. IBM says the team also validated automation on RIKEN's ROQUO GPU supercomputer, reducing the manual transfers that can slow research and introduce errors.

What to watch next

• Whether the workflow continues to improve binding energy accuracy on protein targets that matter to pharmaceutical teams.
• How much of the process can be reproduced by researchers outside the original Cleveland Clinic, RIKEN and IBM collaboration.
• Whether future systems cut the runtime and orchestration burden enough for routine use, not just prize scale demonstrations.

The near term decision value is clear. Organizations evaluating quantum for health research should look less at broad claims of quantum advantage and more at demonstrated hybrid pipelines, resource counts, error handling, and repeatability. This finalist work gives them a richer benchmark to inspect, but it still leaves the central business question open: when the added quantum layer becomes cheaper or more predictive than improving classical chemistry pipelines alone.

Sources

• newsroom.ibm.com
• arxiv.org

Cover photo by Google DeepMind on Pexels, used under the Pexels License.