Computational chemistry and materials science have long relied on advanced computational methods to model molecular systems and predict material properties. Techniques such as density functional theory (DFT) and post-Hartree-Fock approaches (e.g., MP2, coupled-cluster) form the backbone of simulations in both academia and industry. These methods enable accurate modeling of reaction mechanisms, catalyst design, and materials discovery. However, they face intrinsic limitations when addressing systems with strong electronic correlations, such as transition metal complexes or when complex quantum dynamics are involved. The steep computational scaling of more accurate, multi-configurational methods makes them impractical for many chemically and technologically relevant systems. This bottleneck has sparked growing interest in quantum computing as a complementary approach.

Quantum computers promise to represent and process molecular quantum states more efficiently than classical machines. Despite notable progress in hardware, current quantum devices remain constrained by limited qubit counts and imperfect gate operations, while fault-tolerant quantum computers are not yet available. In the meantime, error mitigation techniques seek to recover meaningful signals from ensembles of noisy circuit executions. Although these methods have enabled experimental demonstrations of quantum algorithms for chemistry, they have yet to scale beyond systems accessible to classical high-performance computing. This raises a critical question: can quantum devices deliver a provable and practical advantage for chemistry before full fault tolerance becomes a reality?

One promising approach to address this challenge is the integration of classical high-performance computing with quantum resources, leveraging their respective strengths - a paradigm commonly referred to as quantum-centric supercomputing (QCSC) [1]. For instance, embedding schemes can treat strongly correlated regions of a molecule as an active space on a quantum processor while applying classical methods such as DFT to the remaining degrees of freedom [2]. Another recent development is sample-based quantum diagonalization (SQD), which employs samples from a quantum-prepared electronic wavefunction to perform selected configuration interaction (CI) calculations on a classical supercomputer [3, 4]. This hybrid strategy has already enabled quantum-powered computations to scale beyond the limits of exact diagonalization. Achieving the next milestone, quantum advantage in quantum chemistry, has become a central objective, one that demands a broad, multidisciplinary effort.

Advancing QCSC requires close collaboration between classical and quantum communities [5]. Chemists and materials scientists contribute deep domain expertise, as well as indispensable reference classical state-of-the-art methods, while quantum algorithms researchers design novel formal techniques and implementations. Building bridges between these domains, by identifying synergies, defining benchmarks, and pinpointing near-term use cases, will shape the trajectory of this emerging field.