Classiq, BQP, and NVIDIA have jointly demonstrated an advancement in hybrid quantum-classical simulation for digital twin and computational fluid dynamics (CFD) workloads. The collaboration integrates Classiq’s model-first quantum development platform, BQP’s implementation of the Variational Quantum Linear Solver (VQLS) on the BQPhy platform, and the NVIDIA CUDA-Q platform to enable quantum-ready simulation workflows for high-performance computing (HPC) environments. Utilizing Classiq’s automated circuit synthesis, BQP implemented a VQLS approach that reduces circuit size and optimizes qubit usage, improving the scaling behavior of matrix-based problems common in CFD and digital twin applications and strengthening the viability of hybrid methods within production engineering workflows.
Hybrid Quantum-Classical Simulation Advancement
A joint demonstration by Classiq, BQP, and NVIDIA showcases an advancement in hybrid quantum-classical simulation, specifically for digital twin and computational fluid dynamics (CFD) workloads. The collaboration integrates Classiq’s model-first quantum development platform, BQP’s Variational Quantum Linear Solver (VQLS) implementation on the BQPhy platform, and NVIDIA’s CUDA-Q platform. This allows for the creation of quantum-ready simulation workflows designed for use in high-performance computing (HPC) environments, addressing complex engineering challenges.
BQP leveraged Classiq’s automated circuit synthesis to implement a VQLS-based approach that reduces circuit size, optimizes qubit usage, and lowers the number of trainable parameters compared to traditional methods. These reductions improve the scaling behavior of matrix-based problems common in CFD and digital twin applications. Detailed benchmarks and methodology are available in a technical blog published by BQP: [https://www.bqpsim.com/blogs/vqls-hpc-qc](https://www.bqpsim.com/blogs/vqls-hpc-qc).
This work utilizes NVIDIA CUDA-Q as the execution platform, facilitating integration into existing HPC pipelines used across industry and research. Classiq’s technology transforms high-level models into optimized, hardware-ready quantum circuits, enabling faster algorithm development and cost optimization. BQP has already incorporated these VQLS techniques into its client offerings, ensuring alignment with existing numerical methods and workflow requirements.
Classiq’s Automated Circuit Synthesis
Classiq’s automated circuit synthesis is a key component in a new hybrid quantum-classical workflow developed with BQP and NVIDIA. This technology enables reductions in circuit size, optimized qubit usage, and a lower number of trainable parameters when implementing the Variational Quantum Linear Solver (VQLS). These improvements, compared to traditional quantum linear-solver formulations, enhance the scaling behavior of matrix-based problems common in digital twin and computational fluid dynamics (CFD) applications.
This collaboration leverages Classiq’s synthesis to improve performance within high-performance computing (HPC) environments. Specifically, the work utilizes NVIDIA CUDA-Q as the execution platform, integrating quantum-ready methods directly into existing simulation solutions. BQP has already incorporated these VQLS-based techniques into offerings for clients, aligning them with established numerical methods and solvers used in digital twin, optimization, and simulation.
Classiq’s automated circuit synthesis generates circuits that demonstrate improved scaling characteristics, as illustrated by comparisons to circuits compiled using Qiskit across increasing matrix sizes. The technology transforms high-level functional models into optimized, hardware-ready quantum circuits, enabling faster algorithm development and cost-effective execution. This approach allows enterprises to explore and apply hybrid techniques while maintaining the reliability of their current HPC systems.
This collaboration demonstrates how hybrid quantum-classical approaches can be used today to support demanding engineering workloads. By generating optimized circuits automatically and integrating them into established simulation environments, we enable teams like BQP to incorporate quantum-ready methods directly into the solutions they deliver to customers.
BQP’s VQLS Implementation on BQPhy
BQP has integrated a Variational Quantum Linear Solver (VQLS) implementation onto its BQPhy platform, collaborating with Classiq and NVIDIA. This integration focuses on advancing hybrid quantum-classical simulation for demanding workloads like digital twin and computational fluid dynamics (CFD). Leveraging Classiq’s automated circuit synthesis, BQP’s VQLS approach reduces circuit size, optimizes qubit usage, and lowers the number of trainable parameters compared to traditional methods, improving scaling for matrix-based problems.
This collaborative work utilizes the NVIDIA CUDA-Q platform to support integration into existing high-performance computing (HPC) pipelines. By employing Classiq’s tools, BQP has achieved reductions in circuit size and trainable parameters within its VQLS implementation on BQPhy. These optimizations strengthen the viability of hybrid quantum-classical methods for production engineering workflows, allowing for exploration of these techniques within established systems.
BQP has made a detailed technical blog available, covering the VQLS formulation, benchmarks, and methodology used in the collaboration. This resource, found at https://www.bqpsim.com/blogs/vqls-hpc-qc, provides in-depth information about the advancements achieved through integrating Classiq’s technology with BQPhy and NVIDIA CUDA-Q. BQP currently incorporates these VQLS-based techniques into offerings for its clients.
NVIDIA CUDA-Q Platform for HPC Integration
A collaboration between Classiq, BQP, and NVIDIA has demonstrated advancements in hybrid quantum-classical simulation for demanding workloads like digital twins and computational fluid dynamics (CFD). This work integrates Classiq’s quantum development platform, BQP’s Variational Quantum Linear Solver (VQLS) implementation on the BQPhy platform, and the NVIDIA CUDA-Q platform. The goal is to enable quantum-ready simulation workflows within existing high-performance computing (HPC) environments, offering potential improvements for complex engineering challenges.
The integration leverages NVIDIA CUDA-Q as the execution platform, facilitating the incorporation of quantum methods into established HPC pipelines used by both industry and research. Classiq’s automated circuit synthesis allows for reductions in circuit size, optimized qubit usage, and fewer trainable parameters when implementing VQLS. These improvements enhance the scaling behavior of matrix-based problems common in CFD and digital twin applications, making hybrid methods more viable for production engineering.
BQP has incorporated these VQLS-based techniques into its client offerings, ensuring compatibility with existing numerical methods and solvers. Abhishek Chopra, CEO of BQP, emphasized that this hybrid workflow, executed through the NVIDIA CUDA-Q platform, strengthens the flexibility and scalability of their tools while integrating seamlessly with customers’ established engineering systems. Detailed technical analysis of the methodology and benchmarks is available in a blog post: https://www.bqpsim.com/blogs/vqls-hpc-qc.
Our focus is delivering practical and robust solutions to our clients’ most complex simulation challenges. The hybrid workflow we developed with Classiq and executed through the NVIDIA CUDA-Q platform strengthens the flexibility and scalability of the tools we are deploying today, and it integrates naturally with the engineering systems our customers already rely on.
Abhishek Chopra, CEO of BQP
Scaling and Optimization of Quantum Circuits
Classiq, BQP, and NVIDIA collaborated to advance hybrid quantum-classical simulation for demanding workloads like digital twins and computational fluid dynamics (CFD). Utilizing Classiq’s automated circuit synthesis with BQP’s Variational Quantum Linear Solver (VQLS) on the BQPhy platform, and executed via NVIDIA CUDA-Q, the approach demonstrably reduces circuit size, optimizes qubit usage, and lowers trainable parameters compared to traditional quantum linear-solver formulations. This integration allows for the development of quantum-ready simulation workflows within existing high-performance computing (HPC) environments.
The collaboration’s VQLS-based techniques improve the scaling behavior of matrix-based problems common in CFD and digital twin applications. Classiq-generated circuits, including both width-optimized and depth-optimized versions, were benchmarked against circuits compiled using Qiskit, revealing performance differences related to optimization strategies. BQP has already incorporated these advancements into client offerings, ensuring compatibility with established numerical methods and HPC systems, facilitating practical exploration of hybrid quantum-classical techniques.
Classiq’s platform is designed to transform high-level functional models into optimized, hardware-ready quantum circuits automatically. This automation is key to reducing the complexity of quantum algorithm development, optimizing algorithms for cost and performance, and accelerating the deployment of quantum applications. The technology’s synthesis and memory optimization tools are intended to produce scalable, efficient quantum code, lowering execution costs and enabling broader access to quantum computing capabilities.
Classiq Platform for Quantum Software Development
Classiq, a global leader in quantum computing software, collaborated with BQP and NVIDIA to demonstrate advancements in hybrid quantum-classical simulation. This work focuses on digital twin and computational fluid dynamics (CFD) workloads, integrating Classiq’s model-first quantum development platform with BQP’s Variational Quantum Linear Solver (VQLS) on the BQPhy platform and NVIDIA’s CUDA-Q. This integration aims to create quantum-ready simulation workflows for high-performance computing (HPC) environments, enabling practical application of quantum computing.
Using Classiq’s automated circuit synthesis, BQP implemented a VQLS approach that reduces circuit size, optimizes qubit usage, and lowers the number of trainable parameters compared to traditional methods. This reduction improves the scaling behavior of matrix-based problems common in CFD and digital twin applications. Benchmarking with Qiskit demonstrates how Classiq-generated circuits perform with different optimization strategies, strengthening the viability of hybrid methods in production engineering workflows.
Classiq’s platform transforms high-level functional models into optimized, hardware-ready quantum circuits automatically. This enables faster algorithm development and optimization for cost and performance, offering solutions for both enterprises and researchers. BQP has incorporated these VQLS-based techniques into its client offerings, aligning quantum components with existing numerical methods and HPC systems. A detailed technical blog covering the VQLS formulation and methodology is available at https://www.bqpsim.com/blogs/vqls-hpc-qc.
