Skip to main content
v2026.11,986 entries · CC-BY 4.0
CASRAIAsk CASRAI · AI assistant for researchAsk a research question. Answers come with their sources.Funder rules, compliance, methods and publishing. Answers link the sources they rest on.

Editorial · CASRAI · AI and ML research outputs

Quantinuum and RIKEN Upgrade Reimei to 56-Qubit H2 System

Quantinuum and RIKEN completed upgrading the on-site Reimei trapped-ion quantum computer to 56 qubits for hybrid quantum-AI supercomputing with Fugaku.

Published 16 Nov 2026· 7 minute read

Ask CASRAI · free to try

Ask about this story

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

CASRAI is the reference for research administration — bookmark it for the next question.

On November 10, 2026, Quantinuum and Japan’s national research institute RIKEN announced the completed upgrade of the on-site Reimei quantum computer at RIKEN’s Wako campus to Quantinuum’s 56-qubit System Model H2, creating a three-tier hybrid platform that couples trapped-ion quantum processing with the Fugaku supercomputer and NVIDIA Blackwell GPUs to accelerate quantum-AI research across materials science and biomolecular chemistry.

What happened: Upgrading Reimei to System Model H2

On November 10, 2026, RIKEN held an official ceremony at its Wako campus in Saitama Prefecture to mark the operational commissioning of Quantinuum’s System Model H2 within the Japanese High-Performance Computing and Quantum Technology (JHPC-quantum) platform. The upgrade replaces the earlier 20-qubit System Model H1 with the 56-qubit System Model H2 trapped-ion processor, substantially expanding on-site quantum computing capacity.

The hardware transition completes an installation process initiated in April 2026, when the H2 system was delivered to RIKEN for physical assembly, vacuum pump-down, and calibration. The November announcement confirms that the 56-qubit machine has completed hardware tuning and is actively integrated into RIKEN’s computational network. Alongside the upgrade, Quantinuum (NASDAQ: QNT) and RIKEN signed an expanded Memorandum of Understanding (MoU) to broaden joint research under RIKEN’s “AI for Science” program.

The expanded partnership introduces two strategic commitments: collaborative development with Japanese universities and domestic suppliers toward a fault-tolerant “Made with Japan” quantum computer, and a separate agreement with SoftBank Corp. to evaluate architectures for connecting quantum processors directly to commercial AI data centers across Japan.

Key details: The three-tier hybrid quantum-AI architecture

The upgraded platform at RIKEN establishes a co-processing environment where classical high-performance computing, accelerated GPU AI pipelines, and trapped-ion quantum processors work collaboratively, assigning discrete subroutines to each computing modality:

  • Quantinuum Reimei (System Model H2): Operates 56 fully connected trapped-ion qubits on a Quantum Charge-Coupled Device (QCCD) racetrack architecture, handling mathematically dense calculations involving strongly correlated electron states and deep time-evolution circuits.
  • RIKEN Fugaku Supercomputer: Powered by 158,976 Fujitsu A64FX processor nodes, executing macroscopic physical modeling, multi-scale simulations, and broad data reduction.
  • ROQUO GPU Supercomputer: Equipped with 540 NVIDIA Blackwell GPUs on the GB200 NVL4 platform, performing tensor-network simulations, surrogate model training, and evolutionary AI circuit optimization via NVIDIA CUDA-Q.
  • Quantinuum Nexus: Serves as the overarching workflow management layer, orchestrating asynchronous job queues, circuit compilation, and error handling across all three tiers.
Infrastructure Tier Underlying Architecture Primary Computational Role Key Specifications
Reimei (QPU) Trapped-ion QCCD (racetrack geometry) Strongly correlated electron states and deep time-evolution circuits 56 fully connected physical qubits, high two-qubit gate fidelity
Fugaku (HPC) Fujitsu A64FX (ARM v8.2-A SVE) Macroscopic physical modeling and classical environment simulation 158,976 compute nodes, 442 petaFLOPS peak FP64
ROQUO (GPU/AI) NVIDIA Blackwell GPUs (GB200 NVL4) Tensor-network contraction, surrogate model training, and AI circuit design 540 Blackwell GPUs, CUDA-Q hybrid integration
Quantinuum Nexus Hybrid workflow management layer Asynchronous job orchestration, runtime compilation, and error-mitigation dispatch Cloud and on-premises DAG scheduler with Guppy FT compiler support

Background and context: The JHPC-quantum initiative

The deployment operates under the JHPC-quantum project, funded by Japan’s New Energy and Industrial Technology Development Organization (NEDO) under the Ministry of Economy, Trade and Industry (METI). The program, titled “Research and Development of Quantum-Supercomputers Hybrid Platform for Exploration of Uncharted Computable Capabilities” (Project Leader: Dr. Mitsuhisa Sato), is funded under the Post-5G Information and Communications Systems project (grant JPNP20017).

Simulating complex many-body quantum systems classically encounters exponential computational barriers. In quantum materials research, classical tensor networks encounter severe entanglement bottlenecks during time-evolution simulations. Integrating Reimei with Fugaku and ROQUO mitigates these limitations by combining quantum sampling with accelerated classical heuristics.

Two frontier research projects underway on Reimei illustrate this workflow:

  • Quantum-AI materials simulation: Testing a 6 × 6 two-dimensional spin lattice to map magnetic phase transitions. ROQUO performs tensor-network calculations in weakly entangled regimes, while Reimei executes deep time-evolution circuits. A Transformer neural network trained on Reimei quantum data performs zero-noise extrapolation across the phase space. Preliminary results show the hybrid model achieves higher predictive accuracy than classical-only AI, with potential applications in materials for ultra-high-field MRI and smart power grids.
  • Photochemical biomolecular simulation: Modeling zinc porphyrin, a light-absorbing molecule central to artificial photosynthesis and solar cell engineering. Fugaku simulates the solvent environment, ROQUO runs configuration-interaction calculations and drives evolutionary AI agents, and Reimei searches for critical wavefunctions in strongly correlated molecular orbitals. This structure accelerates complex quantum chemistry simulation workflows without requiring the QPU to represent spectator atoms.

The initiative also aligns with U.S.-Japan scientific collaborations, including the $1 billion Genesis Mission partnership for AI-driven discovery.

Stakeholder reaction and strategic commitments

Leadership across participating organizations emphasized the strategic importance of integrating dedicated quantum hardware with petascale supercomputing and cloud infrastructure.

Dr. Rajeeb Hazra, CEO of Quantinuum, stated: “Japan’s quantum ambitions require advanced capabilities today and a sustained commitment to developing what comes next. The upgrade to Reimei expands what we can explore with RIKEN across electronics, energy and life sciences. Together, we are building on this work to develop a fault-tolerant quantum computer with Japanese universities and suppliers, while exploring with SoftBank how these capabilities can become accessible to industry through Japan’s AI data centers.”

Dr. Mitsuhisa Sato, Division Director of the Quantum-HPC Hybrid Platform Division at R-CCS, noted: “With the upgraded Reimei quantum computer, we will be able to tackle more complex and larger-scale scientific problems that were difficult to address with the previous H1 system, further advancing quantum–HPC hybrid computing through our ongoing JHPC-quantum project. Furthermore, the newly signed Memorandum of Understanding will pave the way for expanding our research into the use of quantum computing for AI for Science, a key initiative being promoted by RIKEN.”

Takenori Kobayashi, Vice President and Head of Product Research & Development Division at SoftBank Corp., added: “Japan has established a strong foundation for hybrid quantum-HPC computing through the NEDO-supported JHPC-quantum project. Our responsibility is to turn these achievements into industrial value. Together with RIKEN and Quantinuum, we aim to advance a Quantum AI Datacenter that connects quantum computing with AI and HPC, making these capabilities accessible to businesses across Japan.”

What the upgrade means for researchers and research offices

The operational launch of the 56-qubit Reimei system introduces practical considerations for computational researchers, facility managers, and university research administration offices.

For research teams, workflows must shift from isolated QPU execution to hybrid co-processing directed acyclic graphs spanning CPU, GPU, and quantum resources. Maintaining computational reproducibility requires systematically logging circuit transpilation parameters, pulse schedules, and classical optimizer seeds. Because the platform uses evolutionary AI to guide circuit synthesis, research groups must establish clear documentation standards for reproducible AI experiments to ensure independent verification.

For sponsored research and compliance offices, multinational quantum deployments intersect with export-controlled research rules. Trapped-ion systems operating with high gate fidelities involve technologies subject to both U.S. Export Administration Regulations (EAR) and Japanese export controls. University compliance administrators managing international scholars or cross-border compute access must verify identity protocols and adhere to deemed export requirements.

Additionally, the “Made with Japan” initiative creates opportunities for academic-industrial technology transfer. Japanese university engineering groups specializing in photonics, cryogenic electronics, and vacuum fabrication can contribute to domestic hardware components, requiring technology licensing offices to define clear patent and data-rights agreements across collaborative consortia.

What happens next: Test User Program and the roadmap to fault tolerance

Following the completion of the initial spin magnetism and biomolecular projects, RIKEN plans to transition Reimei to its formal Test User Program, granting access to selected researchers across Japanese academia and industry via Quantinuum Nexus.

Simultaneously, SoftBank, RIKEN, and Quantinuum will conduct technical evaluations to assess high-speed network connections and security protocols necessary to link quantum hardware with commercial AI data centers, laying groundwork for future enterprise hybrid services.

From a hardware perspective, findings from Reimei’s on-site operation will support Quantinuum’s development roadmap, which targets the 100-logical-qubit Sol system in late 2027 and the universal fault-tolerant Apollo platform in 2029.

Sources

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →