SandboxAQ has announced a collaboration with NVIDIA to drive faster innovation across industries. The company is leveraging the power of NVIDIA DGX to build advanced AI models that support breakthroughs in biopharma, chemicals, materials, finance, cybersecurity, navigation, and medical imaging.
As a member of the NVIDIA Inception program, SandboxAQ is developing a next-generation Large Quantitative Model (LQM) platform. Built on NVIDIA DGX, this platform is designed to solve complex scientific and business problems with more speed, scale, and precision.
With this collaboration, SandboxAQ is improving customer outcomes across several areas.
First, drug, chemical, and materials development is becoming faster. Using NVIDIA DGX, SandboxAQ can reduce discovery cycles from years to weeks. This is done through simulations that replace slow lab experiments, helping teams test and validate new ideas quickly.
Second, the partnership enables the creation of high-quality scientific datasets. By combining chemical and biological simulations, SandboxAQ can detect interactions that were previously hard to identify. These datasets improve model accuracy and reduce false positives.
Third, a new AI Chemist powered by LQMs and NVIDIA DGX is transforming how discoveries are made. This tool can explore millions of chemical pathways automatically, allowing researchers to find new molecules and optimize compounds more efficiently.
The collaboration builds on earlier success. In 2024, the companies achieved an 80x acceleration in quantum chemistry calculations using CUDA-accelerated DMRG. In 2025, they published research showing orbital optimization on a system with 82 electrons and 82 orbitals—double the size of previous simulations.
This joint effort has broad impact. In healthcare, SandboxAQ helps pharma companies speed up preclinical testing. In materials science, it supports the development of sustainable processes and energy storage. In cybersecurity, it enables better modeling and prediction.
The core advantage lies in SandboxAQ’s LQMs. These AI models reflect the laws of science and economics. They provide deterministic and reliable outputs, unlike general-purpose AI models. This makes them ideal for high-stakes environments.
According to Jack Hidary, CEO of SandboxAQ, the collaboration gives customers a clear advantage. “By building on NVIDIA DGX, we help our customers innovate faster and lead with confidence,” he said.
Alexis Bjorlin, Vice President of NVIDIA DGX Cloud, added, “SandboxAQ is setting new standards in AI-native science. With NVIDIA DGX, they can deliver performance at scale and solve real-world problems.”
This partnership highlights how advanced AI and high-performance computing are reshaping R&D. With NVIDIA DGX, SandboxAQ is unlocking new levels of discovery and impact across industries.