A joint team from Cleveland Clinic, Japan’s RIKEN research institute and IBM has been named a finalist for the 2026 ACM Gordon Bell Prize, computing’s most prestigious award for high-performance computing achievement, after simulating a 12,635-atom protein system using a hybrid of quantum and classical computing. It’s described as the largest biologically meaningful molecular simulation achieved with quantum computers to date, and it points at a genuinely practical use for quantum hardware years before most experts expect the technology to reach broad commercial maturity.
What the team actually did
The collaboration combined IBM’s Quantum Heron processors with three of the world’s most powerful supercomputers: Fugaku, developed by RIKEN and Fujitsu; RIKEN’s RUQUO system; and Miyabi-G at the University of Tokyo. The quantum processors handled the electronic-structure calculations, the notoriously difficult quantum-mechanical modeling of how electrons behave across a large molecule, while the classical supercomputers managed the surrounding workflow, a division of labor that plays to each system’s actual strengths rather than asking either to do the whole job alone.
The team didn’t jump straight to 12,635 atoms. The same collaboration published earlier work in May 2026 reporting what was then the first known simulation of a 303-atom protein achieved with quantum computers. In under a year, the group scaled their method roughly 40 times in molecule size while also improving calculation accuracy by a reported 210 times, a combination of scale and precision gains that’s unusual to see move together that fast in computational chemistry, where bigger simulations typically come at the cost of accuracy, not alongside an improvement in it.
Why protein-scale quantum simulation is the actual milestone
Simulating how proteins fold, bind and interact with drug molecules is one of computational chemistry’s hardest and most commercially valuable problems, and it’s exactly the kind of calculation classical supercomputers struggle with at meaningful scale, because the quantum-mechanical interactions between thousands of electrons grow in complexity far faster than the number of atoms involved. Quantum computers are theoretically well-suited to this exact category of problem, since they can represent quantum states natively rather than approximating them, but demonstrating that theoretical advantage on a molecule large enough to resemble something a pharmaceutical researcher would actually study, rather than a toy system built to prove a point, is a different and much harder bar to clear.
That’s the specific significance of 12,635 atoms as a number: it’s large enough to represent genuinely relevant biological structures, the kind of protein complexes involved in real drug-target interactions, rather than a simplified stand-in built purely for a benchmark. A fully automated end-to-end workflow was also part of this round of the research, reducing the manual coordination and data movement between quantum and classical resources that has historically made hybrid quantum-classical computing slow and labor-intensive to run, a practical engineering achievement sitting alongside the headline scientific one.
Who’s actually behind the work
The project sits at the intersection of three different kinds of institution, each contributing something the others don’t have on their own. Cleveland Clinic, one of the largest nonprofit academic medical centers in the United States, brings the biomedical research direction and drug-discovery framing. RIKEN, Japan’s largest comprehensive research institution, operates Fugaku, one of the world’s fastest supercomputers, and contributed both computing infrastructure and quantum-chemistry expertise. IBM supplied the Quantum Heron processors and its quantum-computing research team. That three-way structure, a hospital system, a national research institute and a technology company, reflects how quantum-classical hybrid computing research increasingly gets done: no single institution currently has both the quantum hardware and the domain-specific biomedical expertise to run this kind of project alone.
What happens next
The Gordon Bell Prize winner will be announced at SC26, the International Conference for High-Performance Computing, Networking, Storage and Analysis, running November 15 to 20, 2026 in Chicago. Being named a finalist doesn’t guarantee the win, the Gordon Bell Prize draws finalist teams from some of the most well-resourced computing research efforts in the world every year, and the competition among this year’s finalists hasn’t been publicly ranked ahead of the announcement. What the finalist nod does confirm is that the broader research community has judged this specific quantum-classical hybrid approach to protein simulation as one of the year’s most significant computing achievements, regardless of how the final vote lands in November.
Frequently asked questions
What is the ACM Gordon Bell Prize?
It’s one of the most prestigious annual awards in high-performance computing, presented by the Association for Computing Machinery to recognize outstanding achievement in applying supercomputing to real scientific or engineering problems. The 2026 winner will be announced at the SC26 conference in Chicago in November.
How large was the protein system simulated in this research?
The team simulated a 12,635-atom protein system, described as the largest biologically meaningful molecular simulation achieved with quantum computers to date, up from a 303-atom protein the same collaboration simulated in an earlier paper published in May 2026.
What hardware did the team actually use?
IBM’s Quantum Heron processors handled the quantum electronic-structure calculations, working alongside three supercomputers: RIKEN and Fujitsu’s Fugaku, RIKEN’s RUQUO system, and the University of Tokyo’s Miyabi-G, with classical systems managing the broader computational workflow around the quantum calculations.
For more on the state of quantum and AI computing infrastructure, see our living AI Technology News Today briefing.
