Beyond AI: How Nvidia is helping scale quantum computing

Quantum is set to be the next frontier of computing. But it is an immense undertaking that has yet to be fully realized; researchers continue to face challenges in a multitude of areas — including error correction, scalability, cost and hardware and software complex and capable enough to handle extremely high-performance simulations. 

To support and accelerate scientific exploration into this next phase of computing, Nvidia today launched Quantum Cloud, which allows users to build and test new quantum apps and algorithms in the cloud. 

The microservice is based on Nvidia’s open-source CUDA-Q quantum computing platform, which the company says is used by three-quarters of the companies deploying quantum processing units (QPUs). 

“Quantum computing presents the next revolutionary frontier of computing,” said Tim Costa, Nvidia’s director of HPC and quantum computing. “It’s going to require the world’s most brilliant minds to bring this future one step closer.” 

Breaking down barriers to quantum

The new Nvidia Quantum Cloud is intended to “accelerate scientific exploration,” the company says. Its capabilities include: 

• The Generative Quantum Eigensolver, was developed in collaboration with the University of Toronto. This technology leverages large language models (LLMs) to enable quantum computers to more quickly discover the ground-state energy of a molecule (that is, the most stable configuration). 
• QC Ware Promethium, which tackles complex quantum chemistry problems such as molecular simulation.
• The integration of Classiq with CUDA-Q, which helps quantum researchers generate large programs and analyze and execute quantum circuits.

Nvidia Quantum Cloud intends to “break down the barriers to explore this transformative technology,” said Costa. The goal is to help “every scientist in the world harness the power of quantum computing and bring their ideas closer to reality.”

Nvidia’s quantum ecosystem

Nvidia says it has 160 partners in quantum computing and many top tech companies and quantum companies are incorporating Quantum Cloud into their services — including Google Cloud, Microsoft Azure, Oracle Cloud, IonQ, IQM Quantum Computers, OQC, ORCA Computing, qBraid and Quantinuum.

At HSBC, for instance, researchers designed a quantum machine learning (ML) application that can detect digital payment fraud. This simulated a “whopping” 165 qubits on Nvidia GPUs — typically, projects don’t extend beyond 40 of these fundamental calculating units. 

The researchers were able to overcome challenges with scale thanks to ML techniques implemented with CUDA Quantum and cuTensorNet software on Nvidia GPUs, according to the company. 

Nvidia is further working with nearly two dozen universities to help train the next generation of computer scientists. This involves designing curricula and teaching materials around CUDA Quantum. The company also recently co-sponsored the quantum hackathon QHack, whose winning team — from Indian company Qkrishi — used CUDA Quantum to develop an algorithm to simulate a material critical to designing better batteries.

“Bridging the divide between traditional computers and quantum systems is essential to the future of computing,” Theresa Mayer, VP for research at Carnegie Mellon University, said in a statement. Nvidia’s partnership with her school and others aims to “help students and researchers navigate and excel in this emerging hybrid environment.”

Supporting quantum projects around the world

Nvidia’s CUDA Quantum and other systems are being used for a multitude of other quantum efforts, as well, the company reports. 

These include: 

• The integration of QC Ware’s Promethium quantum chemistry package with Nvidia Quantum Cloud.
• The ABCI-Q project at Japan’s National Institute of Advanced Industrial Science and Technology, one of the largest supercomputers dedicated to quantum computing research. 
• An Nvidia DGX SuperPOD deployment by the Novo Nordisk Foundation. 
• A deployment of CUDA Quantum on NVIDIA Grace Hopper Superchips by Australia’s Pawsey Supercomputing Research Centre. 
• A new CUDA Quantum integration by Classiq. The Israeli startup’s technology helps high-level functional models automatically generate optimized quantum programs so that researchers can expand their work on future algorithms. 
• An effort by ORCA Computing to build and supply a quantum computing testbed for the UK’s National Quantum Computing Centre. This will include an Nvidia GPU cluster using CUDA Quantum.
• An integration of CUDA Quantum into cloud-based platform qBraid’s developer environment. 
• Research by BlueQubit, which found that Nvidia’s technology provides the “fastest and largest” quantum emulations possible on GPUs.