top of page

IBM and RIKEN Achieve Quantum Breakthrough with Fugaku: Largest and Most Accurate Chemistry Simulation Ever

The landscape of high-performance computing is undergoing a transformation. Hybrid systems that merge classical supercomputers with quantum processors are no longer theoretical experiments—they are becoming practical tools for solving some of the most complex problems in chemistry, materials science, and beyond. The recent milestone achieved by IBM and RIKEN, where the Fugaku supercomputer was orchestrated with the IBM Quantum Heron processor, exemplifies this evolution. This collaboration has set a new benchmark for quantum-centric supercomputing (QCSC), demonstrating unprecedented accuracy and scalability in computational chemistry. The Era of Quantum-Centric Supercomputing Quantum-centric supercomputing represents a paradigm shift in computing. Unlike classical high-performance computing (HPC) alone, which relies solely on deterministic algorithms executed on thousands of CPU or GPU cores, QCSC integrates quantum processors into classical workflows. This approach leverages the strengths of both architectures—classical systems handle large-scale deterministic calculations efficiently, while quantum processors excel at solving combinatorially complex subproblems, particularly those involving quantum states or electron distributions. RIKEN and IBM’s recent experiment underscores this principle. By orchestrating Fugaku—a pre-exascale supercomputer with 158,976 chips, each containing 48 cores—with the IBM Quantum Heron processor, the team established a closed-loop workflow that allowed both systems to continuously share data and results in real time. This tight integration is a critical advance over prior sequential hybrid workflows, where quantum and classical computations were performed in isolated stages, often introducing latency and underutilization of resources. Technical Architecture of the Closed-Loop Workflow The closed-loop architecture designed for this experiment ensured maximal efficiency of both the supercomputer and the quantum processor. Classical supercomputers like Fugaku excel at deterministic matrix operations, while quantum processors, such as Heron, handle complex superpositions and entangled states. Orchestrating these two resources requires: Iterative Task Assignment: Dynamic scheduling assigns subproblems to the system best suited to handle them at a given time, minimizing idle cycles. Sample-Based Quantum Diagonalization (SQD): The quantum processor samples the vast space of electron configurations, identifying critical areas for refinement by Fugaku. Real-Time Feedback: Results from the quantum processor are immediately fed back to Fugaku for further classical computation, reducing latency and accelerating convergence toward the final solution. Mitsuhisa Sato, Division Director of the Quantum-HPC Hybrid Platform at RIKEN, highlighted, “This is a very exciting development for hybrid computing. Efficient orchestration is essential when working at this scale.” The workflow maximizes computational throughput and ensures the considerable financial and operational costs of both systems are justified. Solving Complex Chemistry Problems at Scale The immediate application of this QCSC approach was the computation of the electronic structure of a pair of iron-sulfur molecules, compounds critical to a wide range of biological and chemical processes. Traditional methods for solving the electronic structure of complex molecules are limited by the exponential growth of configuration space as the molecule size increases. Quantum computers can sample this space more efficiently, but classical systems are still needed for verification and large-scale integration of results. In this experiment: Quantum Processor Role: Heron explored the vast combinatorial space of electron arrangements. Classical Processor Role: Fugaku refined these sampled states and performed large-scale deterministic calculations to generate accurate electronic structure data. The result was the largest and most accurate quantum chemistry computation executed on a quantum system to date, surpassing previous quantum-only experiments and matching some of the most advanced classical approximations. Accuracy metrics indicate that this combined workflow produces results that are not only scientifically robust but also scalable for future applications. Advantages of Quantum-Centric Orchestration The integration of quantum processors with classical HPC offers several key advantages: Efficiency at Scale: By orchestrating quantum and classical resources in a continuous loop, idle time is minimized. Both Fugaku and Heron operated near maximum capacity throughout the computation. Scalability: The workflow developed for Fugaku is adaptable to cloud HPC environments, suggesting potential for broader deployment across hybrid quantum-classical platforms. Application Versatility: Beyond chemistry, QCSC can be applied to material discovery, cryptography, optimization problems, and simulations of complex physical systems. Reduced Computational Risk: Classical simulation of electronic structures at this scale is computationally prohibitive. Hybrid QCSC approaches allow high-fidelity approximations without requiring prohibitively expensive pure quantum or classical resources. Hiroshi Horii, IBM Senior Manager of QCSC, emphasized the importance of efficiency: “Quantum and classical resources are precious. If either system sits idle, you waste runtime that could solve critical scientific problems. Our closed-loop workflow ensures continuous utilization and maximizes return on investment for both systems.” Implications for the Future of Quantum Computing The IBM-RIKEN collaboration represents a pivotal step toward realizing quantum advantage—the point at which quantum computers outperform classical machines for practical tasks. By demonstrating that quantum processors can seamlessly integrate with some of the world’s most powerful supercomputers, researchers have created a blueprint for tackling problems that were previously infeasible. Looking ahead, researchers are exploring enhancements to the hybrid workflow: GPU Integration: Incorporating GPUs into the orchestration pipeline could accelerate both quantum sampling and classical computation, improving performance and reducing runtime. Algorithmic Development: Advanced quantum-classical algorithms such as SQD are being refined to exploit both architectures more effectively, optimizing task allocation and error correction. Enterprise Applications: Industries with heavy computational needs, including pharmaceuticals, materials science, and energy, can leverage QCSC for rapid prototyping, simulations, and predictive modeling without costly physical trials. Tomonori Shirakawa, senior research scientist at RIKEN, noted, “With all of Fugaku working in a closed loop with Heron, the results were remarkably accurate. As this work scales, quantum advantage is on the horizon.” Broader Context: Investment and Industry Trends The IBM-RIKEN milestone occurs during a period of aggressive investment in AI and HPC infrastructure. Major technology companies are allocating hundreds of billions of dollars toward AI, quantum computing, and hybrid HPC systems. Fugaku itself, as a pre-exascale supercomputer, represents a billion-dollar investment in hardware capable of supporting these hybrid workflows. Global AI infrastructure spending is projected to surpass $660 billion in 2026. Startups in robotics and AI-driven research raised a record $26.5 billion in 2025, reflecting growing confidence in advanced computational platforms. Large industrial and technology players, including Meta, Microsoft, and Google, are actively developing hybrid quantum-classical systems, signaling the broad applicability of QCSC beyond academic research. The combination of investment, infrastructure, and algorithmic innovation positions QCSC as a core component of the next-generation computing landscape. Key Takeaways Feature IBM-RIKEN Milestone Classical Resource Fugaku pre-exascale supercomputer (158,976 chips, 48 cores each) Quantum Resource IBM Quantum Heron processor (on-premises) Workflow Closed-loop, continuous data exchange Algorithm Sample-based Quantum Diagonalization (SQD) Application Electronic structure of iron-sulfur molecules Real-Time Performance Coordinated task assignment minimized idle time Future Enhancements GPU integration, algorithmic optimization, enterprise deployment This achievement highlights that hybrid quantum-classical systems are no longer conceptual experiments. They provide tangible, scalable solutions to computational challenges that were previously intractable. Conclusion The IBM-RIKEN milestone demonstrates that quantum-centric supercomputing is poised to redefine scientific computation. By combining Fugaku’s immense classical processing power with the IBM Quantum Heron processor in a closed-loop workflow, researchers achieved unprecedented accuracy in chemical simulations and established a roadmap for future hybrid architectures. The implications are far-reaching: industries from pharmaceuticals to materials science stand to benefit from faster, more accurate simulations, while the quantum computing community gains a scalable model for integrating quantum processors into real-world HPC environments. As we advance, collaborations like IBM and RIKEN’s underscore the importance of efficient orchestration, workflow design, and algorithmic innovation in achieving quantum advantage. These developments provide a strong foundation for enterprise adoption of QCSC systems and open the door for next-generation research in quantum chemistry, physics, and beyond. For professionals seeking insights into the future of hybrid quantum-classical computing, and the integration of these technologies into practical applications, the work of IBM and RIKEN offers a blueprint for innovation. By combining massive computational scale with quantum-specific algorithms, we are witnessing the emergence of a new era in scientific computing. Read More: Explore further insights from Dr. Shahid Masood and the expert team at 1950.ai on hybrid quantum-classical systems and their transformative potential for industry and research. Further Reading / External References RIKEN and IBM Orchestrate Fugaku with Quantum Processor for QCSC | IBM Quantum Blog IBM and RIKEN Achieve a Major Quantum-Supercomputing Milestone | TipRanks News

The landscape of high-performance computing is undergoing a transformation. Hybrid systems that merge classical supercomputers with quantum processors are no longer theoretical experiments—they are becoming practical tools for solving some of the most complex problems in chemistry, materials science, and beyond. The recent milestone achieved by IBM and RIKEN, where the Fugaku supercomputer was orchestrated with the IBM Quantum Heron processor, exemplifies this evolution. This collaboration has set a new benchmark for quantum-centric supercomputing (QCSC), demonstrating unprecedented accuracy and scalability in computational chemistry.


The Era of Quantum-Centric Supercomputing

Quantum-centric supercomputing represents a paradigm shift in computing. Unlike classical high-performance computing (HPC) alone, which relies solely on deterministic algorithms executed on thousands of CPU or GPU cores, QCSC integrates quantum processors into classical workflows. This approach leverages the strengths of both architectures—classical systems handle large-scale deterministic calculations efficiently, while quantum processors excel at solving combinatorially complex subproblems, particularly those involving quantum states or electron distributions.


RIKEN and IBM’s recent experiment underscores this principle. By orchestrating Fugaku—a pre-exascale supercomputer with 158,976 chips, each containing 48 cores—with the IBM Quantum Heron processor, the team established a closed-loop workflow that allowed both systems to continuously share data and results in real time. This tight integration is a critical advance over prior sequential hybrid workflows, where quantum and classical computations were performed in isolated stages, often introducing latency and underutilization of resources.


Technical Architecture of the Closed-Loop Workflow

The closed-loop architecture designed for this experiment ensured maximal efficiency of both the supercomputer and the quantum processor. Classical supercomputers like Fugaku excel at deterministic matrix operations, while quantum processors, such as Heron, handle complex superpositions and entangled states. Orchestrating these two resources requires:

  • Iterative Task Assignment: Dynamic scheduling assigns subproblems to the system best suited to handle them at a given time, minimizing idle cycles.

  • Sample-Based Quantum Diagonalization (SQD): The quantum processor samples the vast space of electron configurations, identifying critical areas for refinement by Fugaku.

  • Real-Time Feedback: Results from the quantum processor are immediately fed back to Fugaku for further classical computation, reducing latency and accelerating convergence toward the final solution.

Mitsuhisa Sato, Division Director of the Quantum-HPC Hybrid Platform at RIKEN, highlighted, “This is a very exciting development for hybrid computing. Efficient orchestration is essential when working at this scale.” The workflow maximizes computational throughput and ensures the considerable financial and operational costs of both systems are justified.


Solving Complex Chemistry Problems at Scale

The immediate application of this QCSC approach was the computation of the electronic structure of a pair of iron-sulfur molecules, compounds critical to a wide range of biological and chemical processes. Traditional methods for solving the electronic structure of complex molecules are limited by the exponential growth of configuration space as the molecule size increases. Quantum computers can sample this space more efficiently, but classical systems are still needed for verification and large-scale integration of results.

In this experiment:

  • Quantum Processor Role: Heron explored the vast combinatorial space of electron arrangements.

  • Classical Processor Role: Fugaku refined these sampled states and performed large-scale deterministic calculations to generate accurate electronic structure data.

The result was the largest and most accurate quantum chemistry computation executed on a quantum system to date, surpassing previous quantum-only experiments and matching some of the most advanced classical approximations. Accuracy metrics indicate that this combined workflow produces results that are not only scientifically robust but also scalable for future applications.


Advantages of Quantum-Centric Orchestration

The integration of quantum processors with classical HPC offers several key advantages:

  1. Efficiency at Scale: By orchestrating quantum and classical resources in a continuous loop, idle time is minimized. Both Fugaku and Heron operated near maximum capacity throughout the computation.

  2. Scalability: The workflow developed for Fugaku is adaptable to cloud HPC environments, suggesting potential for broader deployment across hybrid quantum-classical platforms.

  3. Application Versatility: Beyond chemistry, QCSC can be applied to material discovery, cryptography, optimization problems, and simulations of complex physical systems.

  4. Reduced Computational Risk: Classical simulation of electronic structures at this scale is computationally prohibitive. Hybrid QCSC approaches allow high-fidelity approximations without requiring prohibitively expensive pure quantum or classical resources.

Hiroshi Horii, IBM Senior Manager of QCSC, emphasized the importance of efficiency:

“Quantum and classical resources are precious. If either system sits idle, you waste runtime that could solve critical scientific problems. Our closed-loop workflow ensures continuous utilization and maximizes return on investment for both systems.”

Implications for the Future of Quantum Computing

The IBM-RIKEN collaboration represents a pivotal step toward realizing quantum advantage—the point at which quantum computers outperform classical machines for practical tasks. By demonstrating that quantum processors can seamlessly integrate with some of the world’s most powerful supercomputers, researchers have created a blueprint for tackling problems that were previously infeasible.

Looking ahead, researchers are exploring enhancements to the hybrid workflow:

  • GPU Integration: Incorporating GPUs into the orchestration pipeline could accelerate both quantum sampling and classical computation, improving performance and reducing runtime.

  • Algorithmic Development: Advanced quantum-classical algorithms such as SQD are being refined to exploit both architectures more effectively, optimizing task allocation and error correction.

  • Enterprise Applications: Industries with heavy computational needs, including pharmaceuticals, materials science, and energy, can leverage QCSC for rapid prototyping, simulations, and predictive modeling without costly physical trials.

Tomonori Shirakawa, senior research scientist at RIKEN, noted, “With all of Fugaku working in a closed loop with Heron, the results were remarkably accurate. As this work scales, quantum advantage is on the horizon.”


Broader Context: Investment and Industry Trends

The IBM-RIKEN milestone occurs during a period of aggressive investment in AI and HPC infrastructure. Major technology companies are allocating hundreds of billions of dollars toward AI, quantum computing, and hybrid HPC systems. Fugaku itself, as a pre-exascale supercomputer, represents a billion-dollar investment in hardware capable of supporting these hybrid workflows.

  • Global AI infrastructure spending is projected to surpass $660 billion in 2026.

  • Startups in robotics and AI-driven research raised a record $26.5 billion in 2025, reflecting growing confidence in advanced computational platforms.

  • Large industrial and technology players, including Meta, Microsoft, and Google, are actively developing hybrid quantum-classical systems, signaling the broad applicability of QCSC beyond academic research.

The combination of investment, infrastructure, and algorithmic innovation positions QCSC as a core component of the next-generation computing landscape.


Key Takeaways

Feature

IBM-RIKEN Milestone

Classical Resource

Fugaku pre-exascale supercomputer (158,976 chips, 48 cores each)

Quantum Resource

IBM Quantum Heron processor (on-premises)

Workflow

Closed-loop, continuous data exchange

Algorithm

Sample-based Quantum Diagonalization (SQD)

Application

Electronic structure of iron-sulfur molecules

Real-Time Performance

Coordinated task assignment minimized idle time

Future Enhancements

GPU integration, algorithmic optimization, enterprise deployment

This achievement highlights that hybrid quantum-classical systems are no longer conceptual experiments. They provide tangible, scalable solutions to computational challenges that were previously intractable.


Conclusion

The IBM-RIKEN milestone demonstrates that quantum-centric supercomputing is poised to redefine scientific computation. By combining Fugaku’s immense classical processing power with the IBM Quantum Heron processor in a closed-loop workflow, researchers achieved unprecedented accuracy in chemical simulations and established a roadmap for future hybrid architectures. The implications are far-reaching: industries from pharmaceuticals to materials science stand to benefit from faster, more accurate simulations, while the quantum computing community gains a scalable model for integrating quantum processors into real-world HPC environments.


As we advance, collaborations like IBM and RIKEN’s underscore the importance of efficient orchestration, workflow design, and algorithmic innovation in achieving quantum advantage. These developments provide a strong foundation for enterprise adoption of QCSC systems and open the door for next-generation research in quantum chemistry, physics, and beyond.


For professionals seeking insights into the future of hybrid quantum-classical computing, and the integration of these technologies into practical applications, the work of IBM and RIKEN offers a blueprint for innovation. By combining massive computational scale with quantum-specific algorithms, we are witnessing the emergence of a new era in scientific computing.


Explore further insights from Dr. Shahid Masood and the expert team at 1950.ai on hybrid quantum-classical systems and their transformative potential for industry and research.


Further Reading / External References

Comments

bottom of page