Nvidia has taken a significant step toward advancing quantum computing with the launch of ABCI-Q, the world’s largest research supercomputer dedicated to quantum-AI development. Located at Japan’s newly established Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT), this initiative is led by the National Institute of Advanced Industrial Science and Technology (AIST) and reflects growing momentum in combining quantum capabilities with high-performance AI systems.
ABCI-Q is built to support hybrid workloads that merge quantum processors with traditional GPU-based computing. The system is powered by 2,020 of Nvidia’s H100 GPUs and relies on the company’s high-bandwidth Quantum-2 InfiniBand networking platform to deliver tightly integrated, large-scale performance. This configuration is designed to support both research and early commercial applications, enabling scientists and engineers to explore practical paths to scalable quantum computing.
At the core of this integration is CUDA-Q, Nvidia’s open-source platform developed to manage and coordinate quantum-classical computing environments. CUDA-Q helps bridge the gap between today’s powerful AI systems and the emerging capabilities of quantum processors, providing a flexible infrastructure for developing and testing algorithms at scale.
The ABCI-Q system is particularly notable for its incorporation of multiple quantum modalities. These include a superconducting qubit processor from Fujitsu, a neutral atom processor from QuEra, and a photonic processor from OptQC. This multi-platform setup allows researchers to experiment with different quantum technologies within a unified, GPU-supported environment—a critical feature as the industry explores which architectures will prove most viable in real-world scenarios.
“Seamlessly coupling quantum hardware with AI supercomputing will accelerate realizing the promise of quantum computing for all,” said Tim Costa, senior director at Nvidia. He emphasized that the collaboration with AIST is expected to drive key advancements in areas like quantum error correction and software development, both of which remain major barriers to practical adoption.
ABCI-Q is not positioned as a quantum computer in the traditional sense. Instead, it functions as a testbed for hybrid architectures and early-stage applications, giving researchers a high-performance environment in which to explore and refine the building blocks of future quantum systems.
For Japan, the supercomputer also represents a strategic investment in next-generation computing. As noted by Masahiro Horibe, deputy director of G-QuAT, the system will help accelerate research into the technical limitations currently holding back quantum computing, particularly around hardware stability, algorithm efficiency, and cross-platform compatibility.
While quantum computing remains in its early stages, Nvidia’s role in ABCI-Q highlights the increasing convergence of AI and quantum research. As demand for computing power surges across fields like drug discovery, energy optimization, and financial modeling, hybrid systems like ABCI-Q may serve as an important bridge—laying the groundwork for more scalable and commercially viable quantum applications in the years to come.
