QuaSi
Revolutionizing computational fluid dynamics with quantum-enhanced neural networks
Hybrid quantum-classical solutions for computational fluid dynamics
We propose hybrid quantum-classical solutions (e.g., PINNs, DeepONets, PINNFormer with quantum submodules) which can remain competitive and strategically valuable by focusing on Quantum-Classical High-Performance Computing (QC-HPC) integration and scalability over time.
By integrating quantum computing into existing HPC workflows, our models serve as drop-in accelerators for specific computational bottlenecks (e.g., spectral transforms, linear solvers, uncertainty quantification). Even partial quantum acceleration of these tasks can offload expensive subroutines from classical clusters, freeing up HPC resources and reducing overall computational costs.
This incremental integration enables parallel co-processing, where quantum and classical components collaborate rather than compete—achieving speedups in multi-step CFD pipelines even if quantum modules are not yet dominant end-to-end.
Transforming scientific computing across sectors
Our quantum PINNs framework revolutionizes single crystal growth modeling with unprecedented precision, accelerating materials discovery for semiconductors and optics.
Learn moreAdvanced quantum-classical models for cryogenic vessel optimization, accurately predicting fluid behavior, heat transfer, and pressurization dynamics.
View DocumentationThe science behind quantum-enhanced simulation
We propose hybrid quantum-classical solution (e.g., PINNs, DeepONets, PINNFormer with quantum submodules) which can remain competitive and strategically valuable by focusing on Quantum-Classical High-Performance Computing (QC-HPC) integration and scalability over time. By integrating quantum computing into existing HPC workflows, our models serve as drop-in accelerators for specific computational bottlenecks (e.g., spectral transforms, linear solvers, uncertainty quantification). Even partial quantum acceleration of these tasks can offload expensive subroutines from classical clusters, freeing up HPC resources and reducing overall computational costs. This incremental integration enables parallel co-processing, where quantum and classical components collaborate rather than compete—achieving speedups in multi-step CFD pipelines even if quantum modules are not yet dominant end-to-end.
Discover how QuaSi can transform your computational capabilities
Aditya Seshaditya
aditya@quasi.digital
Berlin, Germany
+49 1746959355