Quantum, AI Combine to Accelerate Chemistry Discovery

Nvidia and SandboxAQ have combined quantum simulation, classical computing and AI to make the computational chemistry algorithms used to simulate molecules for industry up to 80 times faster.

This allows scientists to perform highly accurate quantitative AI simulations of real-life systems with much greater accuracy than large language models (LLMs) and other AI models can currently do.

The new method uses large quantitative models (LQMs) – a type of pre-trained generative AI model – and a quantum physics algorithm known as density matrix renormalization group (DMRG) accelerated by Nvidia’s CUDA platform for GPUs.  

According to the partners, this combination of the CUDA-DMRG algorithm, Nvidia’s Quantum simulation platform and accelerated computing accelerates these highly accurate calculations more than 80x compared with traditional 128-core CPU computations.

It can also calculate catalysts and enzyme active sites more than double the size that previous methods could. SandboxAQ researchers use these computational results to train AI networks to optimize for the desired treatment or catalyst.

"Advanced computing is opening new frontiers in scientific research,” said SandboxAQ senior staff scientist Martin Ganahl. “Our use of Nvidia technology has allowed us to address some of the most challenging problems in chemistry.

Related:SandboxAQ, Carahsoft Make Quantum Accessible to Public Sector

"We are not only advancing our understanding of material science and chemistry but also paving the way for the next wave of innovations in drug discovery and catalysis to tackle currently untreatable conditions and find safer and cheaper ways to synthesize molecules and materials."

The new computational simulation technique could be valuable in de-risking clinical trials before they happen. For example, the enzyme cytochrome P450s plays a central role in human drug metabolism and is central to understanding drug toxicity but the way it works in the body has been impossible to model up to now.

"AI supercomputing is helping to solve critical problems in the chemical and pharmaceutical industries," said Tim Costa, Nvidia director of high-performance and quantum computing.

"SandboxAQ's use of the Nvidia Quantum platform is facilitating simulations at an unprecedented scale, enabling scientists to rethink what's possible in computational chemistry."