Nvidia Corp. introduced a new orchestration layer for its open-source CUDA-Q platform, aiming to ease a persistent obstacle for developers building quantum applications. The company announced CUDA-Q Logical during IEEE Quantum Week 2026, which began in Toronto. The layer is intended to speed the quantum industry's move toward newer fault-tolerant quantum systems.
CUDA-Q is a software environment developers use to build hybrid applications that run across classical and quantum architectures. It lets them orchestrate workloads spanning traditional processors such as CPUs and GPUs as well as quantum processing units, or QPUs. Nvidia said the industry's recent shift toward fault-tolerant QPUs has created significant problems for developers, and noted that fault tolerance is required for quantum computers to become commercially viable.
Fault-tolerant systems rely on logical qubits, which are groups of physical qubits that work together to correct errors and keep calculations from being corrupted. Nvidia explained that writing software for logical qubits is extremely difficult because the error-correction code they use alters the underlying physical resources an application needs, disrupting the rest of the system. CUDA-Q Logical addresses this by giving researchers a programmable, verifiable environment to simulate and test fault-tolerant quantum systems. Developers can model algorithms, error-correction methods and QPU architectures side by side, identifying an optimal configuration before hardware is built.
Timothy Costa, Nvidia's vice president and general manager of quantum, said quantum computing is maturing rapidly into the era of logical qubits, leaving researchers in need of an open, customizable platform that captures all aspects of fault-tolerant systems. He said CUDA-Q Logical offers the power and flexibility to explore fully integrated, co-optimized systems across qubit types and architectures, and that it sharply shortens the timeline to useful quantum-GPU supercomputing.
Time savings are the new capability's main benefit. Iceberg Quantum, an Australian quantum startup that designs fault-tolerant architectures, used CUDA-Q Logical to model a new architecture for silicon-based qubits developed by Diraq Pty Ltd. The model indicated that 1,000 logical qubits could be created from just 150,000 physical qubits, ten times fewer than Diraq originally estimated.
Fermi National Accelerator Laboratory also used CUDA-Q Logical to assess error-correction strategies, runtime requirements and algorithms across multiple quantum architectures. Its researchers cut the average development cycle for fault-tolerant algorithms from five months to three weeks, roughly seven times faster. Fermilab Chief Technology Officer Anna Grassellino said fault-tolerant quantum computing is the path to new scientific discovery, but reaching it requires co-designing algorithms, error correction, architectures and hardware together, and that her team explored combinations of those resources in three weeks instead of about five months.
Nvidia also discussed broader use of its Quantum-GPU Supercomputing Platform, a cloud service pairing high-performance GPUs with quantum processors for experimental applications. Diraq is using Nvidia's open-source Ising models to calibrate its processors. Anyon Computing and Quantum Machines are using Nvidia's NVQLink networking technology to connect QPU clusters directly to GPU supercomputers, while BlueQubit, IonQ Inc. and Qedema Quantum Computing have integrated their tools with CUDA-Q to deploy hybrid-quantum workloads in production.
More hardware news from TechManNews.
