SandboxAQ to Launch Scientific AI Models on Google Cloud Marketplace for Drug and Materials Research

Advertisement

The rollout will begin with AQCat for materials and catalyst discovery, followed by AQPotency for computational drug discovery, both expected to launch in the third quarter of 2026.

SandboxAQ has announced that it will bring its Large Quantitative Models (LQMs) to Google Cloud Marketplace, enabling researchers to access scientific AI models through conversational AI tools already available on Google Cloud. The rollout will begin with AQCat for materials and catalyst discovery, followed by AQPotency for computational drug discovery, both expected to launch in the third quarter of 2026.

The company said its LQMs are built using real-world laboratory data and scientific equations, allowing researchers to perform advanced scientific calculations without requiring specialized coding or infrastructure. The models will be integrated into Google Cloud's AI ecosystem, where they can be used alongside frontier AI models such as Gemini.

AQCat is designed to calculate adsorption energy, a key measure of how strongly molecules bind to catalyst surfaces. According to SandboxAQ, the model can help research and development teams evaluate materials and catalyst candidates earlier in the discovery process, potentially reducing time and resources spent on computational modeling and laboratory testing.

AQPotency will focus on computational drug discovery by helping researchers assess large libraries of molecules that bind to biological targets, including disease-related proteins. The model is intended to prioritize the most promising candidates for further drug development.

"Bringing our LQMs to Google Cloud's Marketplace will put the rigor of first-principles science directly into the hands of every researcher, in the tools they already use," said Jack D. Hidary, CEO of SandboxAQ.

Brian Goldstein, Vice President of Strategic AI and ISV at Google Cloud, said the collaboration is intended to help healthcare researchers accelerate drug discovery by making scientific AI models more accessible through Google Cloud Marketplace.

The announcement follows SandboxAQ's launch of AQState on June 23, a virtual screening solution for G protein-coupled receptors (GPCRs) accelerated by NVIDIA BioNeMo Agent Toolkit. GPCRs remain one of the largest classes of drug targets, with a 2025 review published in Nature Reviews Drug Discovery reporting that 516 approved drugs target GPCRs, representing 36% of all approved medicines.

SandboxAQ's latest announcement also builds on its partnership with Google Cloud announced in January 2025, which outlined plans to distribute the company's scientific AI models through Google Cloud Marketplace.

Stay tuned for more such updates on Digital Health News