SandboxAQ Unlocks Scientific and Enterprise Innovation with NVIDIA

SandboxAQ has announced a collaboration with NVIDIA to accelerate breakthrough innovations across biopharma, chemicals, advanced materials, financial services, cybersecurity, navigation, and medical imaging. A member of the NVIDIA Inception programme, SandboxAQ is leveraging the NVIDIA DGX Cloud AI platform to develop a state-of-the-art Large Quantitative Model (LQM) platform, fuelling AI-driven scientific discovery. This positions SandboxAQ uniquely to help customers address their most complex and consequential business challenges through LQMs – delivering breakthroughs where traditional models have underperformed – while offering unmatched speed, scalability, and precision.

This collaboration enables SandboxAQ to unlock pivotal advances specifically aimed at transforming customer outcomes, including:

● Up to 4x Faster Discovery in Design-Make-Test Cycles Across Drug, Chemical, and Materials Pipelines: Accelerated by NVIDIA DGX Cloud, SandboxAQ – pioneers of LQMs – transitions laborious, resource-intensive experimental cycles to computational, equation-based simulations, reducing discovery timelines from years to weeks. This enables rapid iteration on molecular and material candidates optimised for multiple target endpoints simultaneously. Customers can validate promising concepts faster and advance key innovations with greater confidence.

● Breakthrough Datasets, Curated with DGX Cloud: SandboxAQ is producing high-fidelity scientific datasets by combining chemical and biological simulations. These methods use equation-based LQM models to uncover interactions between small molecules and complex biological targets previously difficult to detect, such as conformer libraries for generative chemistry and synthetic affinity data for training predictive models. These datasets underpin causal knowledge graphs and more accurate molecular designs, minimising false positives and improving R&D success rates.

● Agentic AI Chemist – A New Era of Autonomous Discovery: SandboxAQ’s AI Chemist coordinates multiple LQMs to transform the scale of R&D. It autonomously explores millions of potential chemical pathways, far beyond what human chemists can analyse, enabling the discovery of novel molecules and the optimisation of compounds for clinical and manufacturing success.

Research Breakthroughs

Today’s announcement builds on previous collaborative efforts between SandboxAQ and NVIDIA:

● 2024 Breakthrough: SandboxAQ and NVIDIA achieved an 80x acceleration in quantum chemistry calculations using CUDA-accelerated Density Matrix Renormalisation Group (DMRG), enabling accurate simulation of enzyme active sites and complex catalysts that were previously computationally infeasible.

● 2025 Breakthrough: In the recently published joint paper, “Orbital Optimisation of Large Active Spaces via AI-Accelerators,” researchers successfully carried out orbital optimisation on a system with 82 electrons in 82 orbitals – more than doubling the simulation scale of previous efforts. This advancement in GPU-accelerated quantum chemistry opens up a realm of molecular simulations previously out of reach, with far-reaching implications for catalysis, material science, and high-dimensional parameter optimisation.

Transformative Impact Across Key Customer Domains

SandboxAQ’s enhanced capabilities deliver strategic value across customer innovation and mission-critical workflows:

● Biopharma and Healthcare: Demonstrated ability to accelerate preclinical pipelines by rapidly generating and refining therapeutic candidates, underpinned by significantly improved predictability of drug efficacy and safety.

● Chemicals and Materials: Facilitates deeper and faster validation of sustainable chemical processes, unlocking applications in carbon and hydrogen utilisation as well as next-generation energy storage technologies.

● Cybersecurity and Strategic Infrastructure: Applies advanced modelling and predictive analytics to improve agility, bolster resilience, and support proactive cybersecurity strategies.

Pioneering How Organisations Can Leverage AI Effectively

SandboxAQ is defining a new category of enterprise AI with its proprietary Large Quantitative Models (LQMs) – a platform built specifically to address massively complex, high-stakes challenges where precision and deterministic outputs are essential. Unlike general-purpose frontier models, SandboxAQ’s LQMs are engineered to reflect the fundamental laws of physics, chemistry, biology, and economics – delivering results that are not only predictive but scientifically robust.

In sectors like drug discovery, SandboxAQ trains LQMs on high-fidelity, domain-specific datasets to markedly improve accuracy, reduce false positives, and shorten the path from hypothesis to therapeutic insight. This science-driven, application-specific modelling approach sets SandboxAQ apart, enabling organisations to generate value in areas where conventional AI methods fall short.

“Our expanded work with NVIDIA accelerates our customers’ ability to innovate and lead in their respective domains,” stated Jack Hidary, CEO of SandboxAQ. “By developing our platform on NVIDIA DGX Cloud and continuing our research collaboration, SandboxAQ will deliver performance and insight that give our customers a clear competitive advantage.”

“SandboxAQ is pushing the boundaries of AI-native science,” said Alexis Bjorlin, Vice President of NVIDIA DGX Cloud. “DGX Cloud provides an AI development platform with the scale and performance needed to power SandboxAQ’s cutting-edge capabilities and deliver real-world impact for organisations addressing society’s most pressing challenges.”