IBM, AMD and Oak Ridge build a platform to fix quantum errors mid-calculation
IBM, AMD, and Oak Ridge National Laboratory have teamed up to solve one of quantum computing’s biggest flaws: errors. Their new platform combines IBM’s 120-qubit Nighthawk quantum computer with ORNL’s Umbra supercomputer, powered by AMD’s Instinct GPUs and EPYC CPUs. The goal is to fix mistakes mid-calculation and test algorithms for fusion fuel, superconductors, and batteries.
Editor, Lazyfounder
30 SEC SUMMARY
- IBM, AMD, and Oak Ridge National Laboratory unveiled a platform combining IBM’s 120-qubit quantum computer with ORNL’s AMD-powered HPC system to correct errors in real time.
- The partnership targets quantum computing’s biggest challenge: errors that disrupt complex calculations like drug discovery and fusion energy simulations.
- AMD’s Instinct GPUs and EPYC CPUs power ORNL’s Umbra system, enabling researchers to tweak qubit signals and refine error correction on the fly.
- Earlier tests by IBM and ORNL simulated magnetic materials and fusion fuel, hinting at breakthroughs in superconductors and batteries.
- The platform’s open workflow lets startups see—and adjust—how error correction works beyond vendor-provided tools.
TABLE OF CONTENTS
KEY HIGHLIGHTS
- IBM, AMD, and Oak Ridge National Laboratory launched a co-design platform integrating IBM’s 120-qubit Nighthawk quantum computer with ORNL’s Umbra HPC system, powered by AMD’s Instinct GPUs and EPYC CPUs.
- Quantum errors occur roughly once every few hundred operations, derailing complex calculations like Shor’s algorithm or molecular simulations without correction.
- The platform lets researchers measure qubits in real time, adjust signal timing, and refine error-correction techniques on the fly.
- Earlier simulations by IBM and ORNL tested magnetic materials and fusion fuel configurations, showing potential for superconductors, batteries, and novel drugs.
- AMD’s hardware could become the standard for large-scale quantum-classical workflows, particularly in energy and materials research.
A new platform to fix quantum errors in real time
IBM, AMD, and Oak Ridge National Laboratory have built a platform to tackle quantum computing’s most stubborn problem: errors. The system links IBM’s 120-qubit Nighthawk quantum computer with ORNL’s Umbra high-performance computing system, which runs on AMD’s Instinct MI355X GPUs and 5th Gen EPYC CPUs.
Why errors derail quantum computing
Today’s quantum computers fail often. Their qubits make mistakes about once every few hundred operations, and those errors pile up fast. Without correction, calculations like Shor’s algorithm for breaking encryption or simulations of molecular structures become impossible. The new platform lets researchers measure qubits mid-calculation, adjust the timing of signals that protect them, and refine error-correction techniques on the fly.
Partners aim for speed and scale
Travis Humble, who directs the Quantum Science Center at ORNL, said the partners are combining ORNL’s supercomputing expertise with IBM’s quantum hardware and AMD’s classical computing chips. The goal is to test new ideas at scale and build a foundation for quantum computers that can handle real-world problems—like designing superconductors, optimizing fusion fuel, or discovering new drugs.
Jay Gambetta, an IBM Fellow and the company’s research director, added that the focus has shifted. "The question used to be whether error correction was possible," he said. "Now it’s about how fast we can make it work." IBM’s Nighthawk processor uses a square-lattice design to enable real-time control, while AMD’s GPUs help fine-tune the correction algorithms.
Early tests point to breakthroughs in energy and drugs
IBM and ORNL have already tested quantum-classical simulations. Earlier this year, they used IBM’s quantum hardware and ORNL’s Frontier supercomputer to model magnetic materials and calculate molecular structures of a promising fusion fuel. Those tests could lead to breakthroughs in superconductors, better batteries, and even new pharmaceuticals.
The new platform adds AMD’s hardware to the mix. Its Instinct GPUs and EPYC CPUs now power the classical side of the stack, which could help standardize AMD’s chips for large-scale quantum experiments—especially in energy and materials research.
Why this matters for quantum startups
This partnership reflects a broader shift in quantum computing: hardware is being designed alongside software to solve specific problems. That’s different from the AI boom, where companies like IBM and CoreWeave focused on scaling cloud resources for generic workloads. Here, the target is clear—error correction—and the tools are built to match.
The platform’s open workflow is a rare gift for startups. Researchers can see raw qubit measurements, adjust signal timing, and watch exactly how error correction works—details usually hidden behind vendor SDKs. That transparency could speed up development of quantum applications in energy and pharma, where simulations of fusion fuel and superconductors are already showing promise.
What this means
Lazyfounder analysis — our interpretation, not reported fact.
This partnership turns quantum error correction from a lab experiment into a practical tool—startups that ignore real-time correction risk building on unstable foundations while those that adopt it could outpace rivals in energy, materials, and drug development.
Key takeaways
- Real-time error correction is now a must for quantum-startups; fragile prototypes will fail in commercial applications.
- AMD’s GPUs and CPUs are emerging as the default hardware for scaling quantum-classical workflows.
- Energy and materials—not abstract algorithms—are quantum computing’s first commercial proving grounds.
- IBM’s open workflow gives startups rare visibility into error-correction mechanics, beyond vendor-provided SDKs.
Related on Lazyfounder
Sources
- SiliconANGLE · 2026-11-05
IBM, AMD and Oak Ridge join forces to launch quantum computing error-correction platform
This story is an original summary drafted with AI by Lazyfounder from the reporting listed above and checked by automated validation. Facts are attributed to their original publishers; sections marked as analysis are Lazyfounder's. Where a source is in another language, facts were machine-translated and quotations are reported, not reproduced. Read the original coverage via the links, and see our AI policy and corrections policy.
About the author
Editor, Lazyfounder
Tarun Mottlia edits LazyFounders, covering Indian startups, funding rounds, AI and product launches. Every story on the site is AI-assisted and checked against its cited sources before publication.
More stories by Tarun MottliaGet the LazyFounder Brief
Startup, funding and AI news in a five-minute read. Join the early-access list.
