New Quanscient and Haiqu algorithm targets scalable fluid simulations on quantum computers – News

Researchers from quantum computing companies Quanscient Oy and Haiqu Inc. today announced that they have developed a new quantum computing algorithm designed to make computational fluid dynamics simulations more practical on quantum hardware.

Developed and tested on IBM Corp.’s largest available quantum computer, the IBM Heron R3, the algorithm reduces the number of qubits required to run complex simulations in computational fluid dynamics, or CFD, on quantum computers, demonstrating a viable path toward future industrial-scale solutions.

CFD is used to model the behavior of air, water and other fluids in applications such as aircraft design, automotive engineering and energy systems. It is a core part of simulation and product testing workflows, but it is also highly demanding computationally and often requires large amounts of processing power and time even on advanced classical supercomputers.

The new approach from Quanscient and Haiqu reduces one of the biggest barriers to running CFD workloads on quantum systems: the amount of hardware resources required.

The algorithm cuts the number of qubits and computational operations needed to perform complex simulations. That provides a more feasible way to execute fluid-flow calculations on today’s limited quantum machines and offers a potential route toward more industrially relevant systems over time.

Core to the approach is what the companies call a one-step simplified LBM, or OSSLBM, framework. The method uses a hybrid quantum-classical approach that simplifies how fluid simulations are mapped onto quantum hardware, making it possible to run nonlinear simulations involving obstacles across multiple steps.

In practical terms, the algorithm allows for moving beyond basic proof-of-concept exercises and closer to simulation problems that better resemble engineering workloads.

Haiqu’s middleware and runtime layer played a key role in making the benchmark possible. Its contributions included reducing circuit depth, improving algorithmic subroutines and applying targeted error-reduction techniques. All that is helping support a multistep workflow that would otherwise be difficult to execute on current quantum devices.

“CFD is one of the most computationally difficult branches of simulation with some of the largest impact on the world’s biggest sectors,” said Valtteri Lahtinen, chief scientist of Quanscient. “Quantum computers offer a future path to simulations that are far more complex than what classical computers can handle, which may allow for the design of more efficient vehicles and aircraft, better energy systems and more.”

The researchers say their work represents a new algorithmic framework that reshapes how fluid simulations are performed on quantum computers. It turns a complicated sequence of calculations into a simpler, more efficient process designed for quantum hardware, the companies say.

The hybrid quantum-classical OSSLBM can be executed on current hardware, outlining a practical path for moving beyond simple linear demonstrations toward more realistic, engineering‑relevant quantum fluid simulations as quantum systems continue to mature.

Image: News/Ideogram