We leverage LQMs derived from AI and Physics based approaches to revolutionize the way chemicals and materials are designed and optimized. Our approach accelerates innovation across key industries including catalyst discovery, battery technology, alloy development and much more!
Our solutions offer a ground breaking alternative to traditional materials discovery by combining data driven insights with high accuracy simulations, helping our customers discover and optimize materials against desired reactivity, synthesizability, kinetics, toxicity, mechanical and electronic properties that push the boundaries of performance, efficiency and sustainability.
Our core differentiated solutions include LLM-based chemical data and information extraction, cloud parallelization, automated high-throughput materials discovery workflows and guided multi-objective generative chemistry coupled with LQMs which leverage Density Functional Theory (DFT), Iterative Full Configuration Interaction (iFCI), Generative AI, Bayesian Optimization and Chemical Foundation Models.
Our chemistry and material applications include:
SandboxAQ’s Large Quantitative Models (LQMs) address the challenge of predicting lithium-ion battery end-of-life (EOL) with unprecedented speed and accuracy. Working with NOVONIX, SandboxAQ developed LQMs trained on five years of ultra-high precision coulometry data, achieving predictions in days rather than months. Using physics-based simulations and NOVONIX datasets, LQMs model electrochemical patterns, enabling predictions with 35x greater accuracy and 50x less data than traditional AI models. This innovation reduces testing time by 95%, paving the way for faster commercialization of next-gen batteries and lowering R&D costs while supporting eco-friendly energy storage solutions. [Scientific Manuscript]
SandboxAQ, partnering with AWS, Intel, and Accenture, developed a massively scalable quantum chemistry solution to tackle PFAS pollution [AWS blog]. Turning the AWS cloud into a massively distributed supercomputer for chemistry, the platform enabled near-exact simulation of bond breaking in toxic PFAS molecules for the first time in the history of chemistry, utilizing over one million vCPUs [Scientific Manuscript]. This cloud-native approach enables accurate simulation of complex molecules, which is essential for designing scalable PFAS remediation and more sustainable PFAS replacements. The breakthrough in simulation scale—achieving an unprecedented reduction in cost and time—supports global remediation efforts and opens possibilities for applications in green chemistry, battery materials, and sustainable agriculture.
SandboxAQ, partnering with DIC, AWS collaborated to perform large-scale high-accuracy quantum chemistry using QEMIST Cloud and simulated the largest organometallic catalyst computed at near-exact accuracy to date. Accurate quantum chemistry data is always desired in computational catalyst design, but is typically impractical to generate for all but small model systems due to the computational complexity involved in the simulation of the electronic structure of chemical systems. The present results indicate that our technology breaks this limitation, making it possible to perform highly accurate computations on catalysts at the industrially relevant molecular scale. High-accuracy quantum chemical data resulting in LQM’s for chemistry catapults AI-driven materials design and drug discovery to achieve deep impact at scale!
[AWS blog] [DIC Press Release]
SandboxAQ in partnership in collaboration with Nvidia combined its large quantitative models (LQMs) with the Nvidia CUDA-accelerated Density Matrix Renormalization Group (DMRG) algorithm. The combined technologies makes it possible to perform highly accurate quantitative AI simulation of real-life systems that go beyond what large language models and other AI models can currently do. These have resulted in computing speeds of more than 80x for these simulations with Nvidia’s technology compared to computing on traditional 128-core central processing units. The team's efforts have enabled unprecedented calculations for complex biochemical systems, which include transition metal metalloenzymes. Such metal-containing catalysts are crucial in numerous industrial and biological processes, playing an essential role in facilitating chemical reactions. These powerhouses of energy conversion are vital for many industries, including medicine, energy, and consumer products. They accelerate chemical reactions, lowering the energy required and making processes more efficient and sustainable. Understanding and optimizing these catalysts is essential for addressing global challenges, such as clean energy production and environmental sustainability. This work is captured in a paper [Scientific Manuscript] co-authored between SandboxAQ and Nvidia.
