The Signal
Fermilab, IBM and NVIDIA plan modular quantum platform
Fermilab, IBM and NVIDIA plan a platform combining IBM quantum computers with NVIDIA accelerated computing. Researchers would model links between quantum modules before Fermilab adds its own superconducting cavity hardware in a later phase.

A joint design effort
A proposed Fermilab–IBM–NVIDIA platform would colocate quantum computers with accelerated-computing systems, giving researchers a place to develop and test modular quantum computing approaches. The three organizations announced their plans on November 16, 2026, describing the effort as aimed at advancing quantum and classical computing for scientific research.
Source newsroom.ibm.com
The initiative builds on Fermilab’s work with IBM through the Superconducting Quantum Materials and Systems Center, a U.S. Department of Energy research center. The partners propose to combine quantum hardware, accelerated computing and scientific expertise to investigate how multiple quantum processors could be connected and operated together.
Testing links between modules
The proposed initial setup would bring IBM quantum computers together physically with NVIDIA DGX B200 accelerated-computing systems. Fermilab would lead engagement with high-energy and nuclear physics researchers and coordinate software workflows and researcher access through its HEPCloud infrastructure.
Researchers would use the platform to develop and test quantum algorithms, error correction, superintelligence-assisted optimization and real-time quantum–classical control. The initial work would model and emulate communication between modules, examining how link errors, latency and feedback affect performance.
Those studies are intended to inform the development of physical quantum interconnects by the SQMS Center and IBM. The underlying modularity challenge is how to connect processors so they can work together on calculations beyond the capacity of a single device, the Fermilab release said.
Applications and partner roles
The planned research would include shared software and benchmarks for problems in particle interactions, neutrino physics and nuclear simulations. Fermilab said the work would translate scientific needs into requirements for future quantum computing systems, while bringing high-energy and nuclear physicists into decisions about algorithms and architectures.
IBM would contribute superconducting quantum processors, software, application development and quantum networking expertise. NVIDIA would bring accelerated computing and superintelligence technologies for decoding, simulation, optimization and real-time feedback, while Fermilab would provide cavity and interconnect research and scientific leadership.
A later hardware phase
In a later phase, Fermilab’s SQMS Center intends to integrate its own ultracoherent superconducting cavity hardware into the platform. Researchers would also investigate cavities as quantum memories, entanglement buffers and resources for bosonic error correction.
The organizations describe the longer-term aim as developing technology, software and a scientific user community for future scalable quantum computing facilities.
Sources
Companies in this story
Related stories
- NVIDIA releases QEC decoder, claims up to 13,000× speedup
- Yonsei Installs IBM’s Latest Quantum Nighthawk Processor at Its Quantum Computing Center
- IBM plans 520-qubit Nighthawk r3 and linked system for 2027
- IBM Puts Quantum System Two Online in Chicago
- IBM Nighthawk Prepares a Laughlin-Like Chern Insulator