NVIDIA, IBM and Fermilab Plan Modular Quantum Computing Platform

What happened
On November 16, 2026, NVIDIA, IBM, and Fermilab announced a plan to create a modular quantum-classical computing platform. The initiative proposes to physically co-locate IBM quantum processors with NVIDIA accelerated computing systems at Fermilab. The goal is to develop and test scalable quantum error correction methods and quantum-classical algorithms, focusing on scientific problems in high-energy and nuclear physics. The project is currently at the proposal stage and will involve collaboration on software, hardware, and workflow integration.
Key details
- IBM Quantum Nighthawk processors, supporting up to 120 qubits and 7,500 two-qubit operations, will be used.
- Fermilab will coordinate researcher access and software workflows through HEPCloud.
- The platform will study linked quantum modules and interconnects, including error rates and latency.
- Slurm 26.11 will add native support for quantum processing units, enabling shared scheduling of quantum hardware.
Why it matters
Researchers in high-energy and nuclear physics will gain access to a new platform for developing and testing quantum error correction and scalable quantum algorithms, potentially reducing the time required for complex computations. Quantum hardware providers and supercomputing centers will be able to manage and schedule quantum resources more efficiently, thanks to new software integrations. This initiative creates a pathway for future Department of Energy quantum computing user facilities and may accelerate the development of fault-tolerant quantum systems.
