NVIDIA Strikes Quantum Supercomputing Partnership Agreements

NVIDIA has formed partnerships with quantum computing centers in Germany, Japan, and Poland, which will each adopt the company’s open-source NVIDIA CUDA-Q platform to power their respective quantum processing units (QPUs).

The CUDA-Q open-source platform is designed for integrating and programming QPUs, graphics processing units (GPUs), and central processing units (CPUs) in one system through a unified and open programming model. According to NVIDIA, CUDA-Q enables GPU-accelerated system scalability and performance across heterogeneous QPU, CPU, GPU, and emulated quantum system elements.

The platform is to be deployed at the Poznan Supercomputing and Networking Center (PSNC) in Poland in collaboration with London-based ORCA Computing. The PSNC’s high-performance computing data center is equipped with two ORCA PT-1 Series quantum photonic systems in conjunction with NVIDIA GPU-based clusters. The integration of CUDA-Q at PSNC is expected to accelerate work in a range of scientific and application areas including biology, chemistry, and machine learning.

NVIDIA’s CUDA-Q open-source platform for quantum computing applications will see deployment across three sites in Poland, Japan, and Germany, with each using separate quantum processing units. Courtesy of NVIDIA.

In March, ORCA and NVIDIA reported the integration of the CUDA-Q platform with an ORCA PT-1 QPU. The companies demonstrated an algorithm for image generation using a hybrid generative adversarial network approach. According to ORCA, the output of the quantum processor was fed to neural networks running on GPUs to produce higher quality data.

Meanwhile, in Japan, CUDA-Q has been integrated into the National Institute of Advanced Industrial Science and Technology's (AIST’s) ABCI-Q, which uses the NVIDIA Hopper architecture. The supercomputer is powered by more than 2000 NVIDIA H100 Tensor Core GPUs in 500+ nodes interconnected by the NVIDIA Quantum-2 InfiniBand. The site is also set to deploy a neutral-atom quantum computer from QuEra Computing in 2025.

The QPU integrated with ABCI-Q will enable researchers at AIST to investigate quantum applications in AI, energy and biology, utilizing Rubidium atoms controlled by laser light as qubits to perform calculations.

In addition, Germany’s Jülich Supercomputing Centre (JSC) at Forschungszentrum Jülich will install IQM’s Spark quantum computer in July to complement its JUPITER supercomputer as part of the JUNIQ quantum computing infrastructure. The 5-qubit system, based on superconducting technology, is designed for basic experiments and applications in teaching at universities and research institutes. In addition to CUDA-Q, the system will also include the NVIDIA GH200 Grace Hopper Superchip designed for artificial intelligence and high-performance computing applications.