NVIDIA has unveiled its NVQLink high-speed interconnect, an open system architecture for tightly coupling the extreme performance of GPU computing with quantum processors to build accelerated quantum supercomputers. NVQLink defines an open, modular reference architecture that supports multiple quantum modalities and enables real-time interaction between classical GPU resources and quantum processors through a “real-time hub.” By providing fast, scalable communication across quantum and classical systems, the hub supports tightly coupled hybrid applications such as machine learning, simulation and optimization. The hub also supports control tasks such as quantum error correction, decoding and calibration which will be essential for future fault tolerant quantum computers. NVIDIA NVQLink high-speed interconnect lets quantum processors connect to world-leading supercomputing labs, allowing researchers to build hybrid quantum-classical systems to accelerate next-generation applications in chemistry and materials science. Courtesy of NVIDIA. According to launch partner Infleqtion, NVQLink provides a standardized, low-latency connection between quantum processors and GPU-accelerated systems, enabling true real-time hybrid computing. Infleqtion plans to host NVQLink-enabled Sqale system at the Illinois Quantum and Microelectronics Park. Scaling quantum computers requires logical qubits and fast quantum–classical feedback loops (for calibration, decoding, and error correction). Historically there hasn’t been a single, high-performance way to couple QPUs to AI supercomputers — NVQLink provides that missing interface, Infleqtion said. It supports real-time orchestration of CPUs, GPUs, and QPUs, so quantum programs and AI programs can operate as one system. Under the hood, NVQLink connects quantum control systems to appropriately sized GPU servers and uses an optimized network (RDMA over Ethernet) for deterministic, low-latency data movement— microsecond-scale transfer in typical configurations, which fits comfortably into our expected quantum error correction clock cycle time. NVIDIA's CUDA-Q software platform exposes real-time callbacks across NVQLink, so developers can implement cycle-by-cycle control and hybrid algorithms without custom plumbing. The technology allows quantum processors to connect to supercomputing labs including Brookhaven National Laboratory, Fermi Laboratory, Lawrence Berkeley National Laboratory (Berkeley Lab), Los Alamos National Laboratory, MIT Lincoln Laboratory, Oak Ridge National Laboratory, Pacific Northwest National Laboratory, and Sandia National Laboratories. Researchers from these sites were closely involved in the development of NVQLink. Researchers and developers can access NVQLink through its integration with the NVIDIA CUDA-Q software platform to create and test applications that seamlessly draw on CPUs and GPUs alongside quantum processors, helping ready the industry for the hybrid quantum-classical supercomputers of the future. Partners contributing to NVQLink include quantum hardware builders Alice & Bob, Anyon Computing, Atom Computing, Diraq, Infleqtion, IonQ, IQM Quantum Computers, ORCA Computing, Oxford Quantum Circuits, Pasqal, Quandela, Quantinuum, Quantum Circuits, Inc., Quantum Machines, Quantum Motion, QuEra, Rigetti, SEEQC, and Silicon Quantum Computing — as well as quantum control system builders including Keysight Technologies, Quantum Machines, Qblox, QubiC, and Zurich Instruments.