Overview
- NVIDIA announced CUDA‑Q Logical on Monday, September 14, 2026, and published reference implementations and the Sandia QUOPS benchmark on GitHub for open use.
- CUDA‑Q Logical is a programmable orchestration layer that lets researchers design, verify and simulate error‑corrected logical qubits across different error‑correction schemes and QPU types.
- Fermi National Accelerator Laboratory, among other early adopters, reported big productivity gains from the tooling, with Fermilab saying it cut fault‑tolerant prototyping from five months to three weeks.
- In December 2024, Infleqtion used the same logical‑qubit tooling to run a materials‑science experiment on Sqale’s neutral‑atom processor and observed a measurable drop in logical error rates.
- NVIDIA says additional CUDA‑QX error‑correction libraries and NVQLink for low‑latency QPU–GPU coupling remain in development, which could enable tighter hybrid quantum–GPU workflows for drug discovery, materials research and other compute‑heavy domains.