NVIDIA has expanded its open-source CUDA-Q platform with CUDA-Q Logical, a compilation and orchestration layer for fault-tolerant quantum computing. The company detailed the release on September 14, 2026 at the World Congress on Quantum Software Solutions.
An era of logical qubits, standardised
CUDA-Q Logical gives researchers a programmable, verifiable framework for developing fault-tolerant applications in drug discovery, materials science and financial modelling. It streamlines the codesign of algorithms, error-correction codes, hardware architectures and QPUs — Fermilab says it cut the design cycle for a new fault-tolerant architecture from five months to three weeks.
Partner ecosystem lines up at launch
Adopters at launch include Fermilab, Infleqtion, IQM Quantum Computers, QCDesign, Quantum Motion and Sandia National Laboratories. NVIDIA also announced QUOPS, a benchmarking framework co-developed with Google, IBM and Quantinuum. "Quantum computing is maturing into an era of logical qubits, and researchers need an open, customisable platform capable of representing all aspects of a fault-tolerant system," said Timothy Costa, VP and GM of quantum at NVIDIA.
The software side of the physical qubit race
The release complements this week's Anyon Computing NVQLink control-plane launch and slots alongside earlier partnerships such as the Altera–Riverlane error-correction integration and IQM's Lumi-IQ logical qubit rollout. Together they are building the missing software abstraction that fault-tolerant quantum needs to leave the lab.
Reporting based on coverage from NVIDIA, HPCwire and The Quantum Insider.