Quantum is coming, just keep waiting. . .
https://www.barrons.com/articles/ibm-quantum-computing-research-nvidia-amd-0dd74344
By: Mackenzie Tatananni
Updated Jan 29, 2026, 12:26 pm EST / Original Jan 29, 2026, 12:01 am EST
It might seem natural to pit the capabilities of quantum computers—often touted as the next big technology—against today’s supercomputers. But scientists have a different, more collaborative vision for the future.
Rather than outright replacing classical machines, quantum systems will likely be built on top of existing architecture to enable more powerful computations.
This perspective was captured in new research from International Business Machines, which showed how classical graphics processing units, or GPUs, from leading chip makers could work alongside quantum processors to execute problems faster.
“It’s important for the world to see that quantum computers aren’t just these things that [will] replace your computers,” Jerry Chow, IBM’s chief technology officer of quantum-centric supercomputing, said in an interview with Barron’s. “They’re really part of the entire computing infrastructure.”
Two papers co-authored by IBM researchers detailing the results were posted to Cornell University’s arXiv, a research-sharing platform for papers that have yet to be peer-reviewed.
The first paper showed how GPUs from Advanced Micro Devices, contained within the Frontier supercomputer at Oak Ridge National Laboratory, could be combined with IBM’s Heron processor to model complex chemical systems and provide a 100-time speedup over central processing units, or CPUs.
“It’s a notoriously difficult problem for classical computers,” Chow explained. “But we know now, there’s a part of these chemistry problems which are best handled on the quantum computer.”
Follow-up work with Riken, a Japanese research institute, was showcased in the second paper. The study showed how the same chemical systems could be modeled on IBM quantum computers for an additional 20% increase in performance when using a new algorithmic approach on Nvidia chips.
“Overall, they’re all supporting the same narrative, that the future of computing is quantum-centric,” Chow said. As he sees it, it’s the next logical step in a continuous evolution of technology.
Classical GPUs have already proven their prowess at one type of math. The chips perform calculations by breaking large problems into thousands of simpler tasks that are processed simultaneously, through a method called parallel processing.
Meanwhile, quantum processors harness the properties of quantum mechanics, which makes them best suited for complex modeling tasks. This explains why most quantum research consists of some kind of modeling problem—and why quantum is expected to have an outsize impact in the areas of dr-g discovery and materials science.
Classical processors are “the technology behind everything that we see today with language models and training and inference and so forth,” according to Chow.
It’s impossible to rule out a distant future “where everything is all based off the same kind of technology,” he added. “But at least from what we’ve seen with how supercomputing has evolved in the last 10 to 15 years, it’s all about composable pieces.”
Jay Gambetta, who oversees IBM’s research effort, shared a similar view in an earlier interview with Barron’s. “We’re imagining a heterogeneous accelerating framework that connects quantum and classical compute,” Gambetta explained during a tour of IBM’s Thomas J. Watson Research Center in October.
Much of IBM’s past research has incorporated hybrids of quantum and classical computing. In September, computers running on IBM’s Heron processor worked alongside bit-based machines to perform a bond trading problem.
One month earlier, IBM and AMD unveiled a formal collaboration to develop quantum-centric supercomputers. The partners indicated they were exploring ways for AMD chips to control errors on IBM’s quantum processors, which could advance IBM’s efforts to develop fault-tolerant quantum computers by the end of the decade.
Crucially, the latest results demonstrate that real-world applications of quantum computers are just a step away.
“We have a number of partners who already have clusters of supercomputers that they know how to access,” Chow said. “To them, it’s like, ‘Now I have the keys to a brand-new car, let me see what I can do with it and how I can work it in with what I’ve been doing.’”
IBM shares climbed 6.8% on Thursday, bolstered by strong quarterly earnings. The tech-heavy Nasdaq Composite fell 1.9%. So far this year, IBM has gained 6.1%, outstripping a 0.7% gain for the index.