Japan’s National Institute of Advanced Industrial Science and Technology (AIST), in collaboration with NVIDIA, has launched ABCI-Q, a new research-focused supercomputing system designed to support large-scale hybrid quantum-classical computing. The system is hosted at the newly established Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT) and is powered by 2,020 NVIDIA H100 GPUs interconnected using NVIDIA’s Quantum-2 InfiniBand platform. The hybrid infrastructure integrates CUDA-Q, an open-source platform for orchestrating large-scale quantum-classical computing, enabling researchers to simulate and accelerate quantum applications at scale.
ABCI-Q supports hybrid workloads by combining GPU-based simulation with physical quantum processors from multiple vendors. The system incorporates superconducting qubits from Fujitsu, neutral atom qubits from QuEra, and photonic qubits from OptQC. This modular, multi-qubit architecture allows for testing quantum error correction, algorithm development, and co-design strategies critical to future quantum systems. The ABCI-Q infrastructure serves as a testbed for evaluating quantum-GPU workflows and advancing practical use cases across multiple hardware modalities.
The project aims to support quantum research under Japan’s broader quantum strategy while showcasing the role of accelerated classical hardware in enabling quantum advancement. According to AIST, ABCI-Q will help scientists address key challenges in hybrid quantum-classical computing and refine foundational technologies for scalable, fault-tolerant systems. NVIDIA’s involvement underscores the strategic use of its GPU ecosystem in quantum algorithm testing and hardware integration.
Full details are available in the official announcement from NVIDIA here.
May 19, 2025
Leave A Comment