Osaka University and Fixstars Shatter the 40-Qubit “Glass Ceiling” in Quantum Chemistry Simulation
Iterative Quantum Phase Estimation IQPE
Researchers from Fixstars Corporation and Osaka University’s Center for Quantum Information and Quantum Biology (QIQB) have officially broken through the long-standing “glass ceiling” of 40-qubit classical quantum simulation, a significant accomplishment for the fields of computational chemistry and quantum informatics. The collaborative team successfully completed one of the biggest state-vector-based simulations of quantum chemical circuits ever documented by utilizing the enormous capability of 1,024 NVIDIA H100 GPUs. This innovation creates a new high-fidelity standard for the creation of algorithms intended for upcoming Fault-Tolerant Quantum Computers (FTQC) and was carried out on the ABCI-Q system at Japan’s National Institute of Advanced Industrial Science and Technology (AIST).
The Challenge of the 40-Qubit Barrier
The 40-qubit threshold has long been a significant obstacle for conventional supercomputers trying to do state-vector simulations. Each qubit doubles the quantum state data needed, making memory exponentially increasing. Handling so much data requires massive hardware and an innovative parallel computing architecture that can overcome inter-GPU communication limitations.
Because the complexity of syncing data between processors starts to outweigh the speed advantages of parallelization, standard simulations frequently stall at high qubit counts. To get around this, Fixstars offered specific performance profiling and tuning knowledge to synchronize the 1,024 H100 GPUs on the ABCI-Q cluster. The team was able to go beyond the conventional boundaries of classical simulation with its custom architecture, which guaranteed that the enormous hardware footprint could be used effectively.
Methodology: The Power of IQPE
Iterative Quantum Phase Estimation (IQPE) is the technical foundation of this accomplishment. Compared to traditional Quantum Phase Estimation (QPE), IQPE is a crucial subroutine in quantum chemistry that extracts exact energy eigenvalues from a system while requiring a substantially smaller number of supplementary qubits.
The researchers used the “chemqulacs-gpu” simulator, an enhanced version of the Qulacs library designed especially for high-performance GPU clusters, to run these intricate simulations. Once large-scale, error-corrected hardware becomes available, IQPE is usually regarded as the best option for industrial applications in domains like drug discovery and materials research due to its resource efficiency. A crucial testbed for the software that will eventually run on upcoming quantum processors is provided by the ability to simulate these circuits classically now.
Molecular Milestones: From Water to Iron-Sulfur Clusters
The simulation achieved two noteworthy molecular complexity benchmarks that highlight the scope of this advancement:
- H2O Molecule: The group demonstrated the ability to manage deeper circuits and more intricate Hamiltonian terms by simulating a 42-spin-orbital system for a water molecule.
- Iron-Sulfur (Fe2S2) Cluster: A 41-qubit pure circuit benchmark for an iron-sulfur cluster was attained by the researchers.
The intricate electrical architecture of iron-sulfur clusters make modeling them infamously challenging for classical computers. Nonetheless, these clusters are extremely significant biologically because they are essential to processes like photosynthesis and nitrogen fixation. Through the effective simulation of these systems, the researchers have increased the number of tools accessible for testing the gate sequences and Hamiltonians that will propel future catalyst and sustainable material discoveries.
Strategic Importance: “De-Risking” the Future
The treatment of software optimization as a fundamental pillar for quantum utility is a growing change in the quantum roadmap that is highlighted by this collaboration between academics and industry. Having traditional “gold standards” is crucial as the industry approaches the era of Fault-Tolerant Quantum Computing. When quantum subroutines are eventually scaled up to hardware, these criteria guarantee that they operate precisely as anticipated.
Today, researchers can successfully “de-risk” their software stacks by simulating 42-qubit circuits on conventional supercomputers. This ensures a smooth and successful transition to quantum hardware by enabling industrial instruments for molecular screening to be thoroughly tested and validated prior to the arrival of large-scale, fault-tolerant systems.
Looking Ahead to GTC 2026
The outcomes of this groundbreaking research will be showcased during the GTC 2026 conference (Session P81339). The performance tweaking and parallelization techniques that enabled the ABCI-Q system to overcome the 40-qubit barrier will be thoroughly explained technically in this talk.
The achievements of Fixstars and Osaka University serve as a reminder that classical innovation is paving the way for quantum advantage. The promise of quantum-driven drug discovery and materials science is getting closer to reality as researchers continue to push the limits of what is feasible on classical GPUs, closing the gap between present simulations and future quantum applications.
Interested parties should refer to Fixstars’ technical announcements and Osaka University’s official research news for more information. The ability of high-performance computers to reveal the mysteries of the quantum world is demonstrated by this work.
