SandboxAQ announced that AQCat is now generally available on Claude Science through Model Context Protocol (MCP), letting any researcher tap the company’s Large Quantitative Model (LQM), an AI model trained to predict the physics of materials, simply by asking questions in plain English. AQCat  identifies promising catalysts with accuracy approaching the field’s most trusted methods, but up to 20,000 times faster, and it runs entirely through natural-language prompts with no coding required.

Catalysts are the hidden engine of the physical economy. They are the materials that speed up chemical reactions and make them practical at industrial scale, and they are essential to more than 80% of all manufactured goods and over 90% of all commercially produced chemicals, from fertilizer and fuel to fabric, glass, and plastics. Yet finding new and better ones has remained slow and expensive. Conventional lab methods typically test fewer than 100 candidate materials per week, a pace that keeps most progress limited to small, incremental gains.

AQCat lets any research team evaluate catalysts at a scale that once required specialized staff and hard-to-access computing power. It runs the first and most decisive step in the process: calculating adsorption energy, a measure of how strongly molecules stick to a catalyst’s surface, which is the key early signal of whether a material will work. Teams can rank thousands of candidates up front and focus their modeling and lab budgets only on the most promising ones.

Marketing Technology News: MarTech Interview with Mark Listes, CEO @ Pendulum Intelligence

AQCat’s accuracy comes from physics that other AI models leave out. Density Functional Theory (DFT), the long-standing gold standard for these calculations, is highly accurate but slow and costly, while faster machine-learning models often ignore magnetism altogether. AQCat is “spin-aware,” meaning it accounts for magnetic behavior, bringing abundant, low-cost but magnetically complex metals like iron, cobalt, and nickel within practical reach. SandboxAQ trained it on AQCat25, a public dataset of 13.5 million high-fidelity DFT calculations spanning 47,000 intermediate-catalyst systems and every industrially relevant element.

“In catalysis research, a key bottleneck has been the number of surfaces you can realistically evaluate within compute and scientist resource constraints. Recovering DFT-quality adsorption energies at a faster speed changes the scale of questions my group can ask, and it opens up design spaces that were previously out of reach. Putting models behind a natural-language interface lowers the barrier for students and collaborators to participate in computational work,” said Dr. Joe Gauthier, Assistant Professor of Chemical Engineering, Texas Tech University.

“Adsorption energy is the critical first calculation in most catalyst screening workflows, because it tells you how strongly a molecule binds to a surface and helps you prioritize what to study next. Many AI models overlook the magnetic physics that matter in earth-abundant industrial catalysts. We built AQCat to be spin-aware from the ground up, so it can recover near-DFT accuracy for magnetically complex systems, including iron, cobalt, and nickel, while running thousands of times faster than traditional approaches. Making it generally available on Claude means any researcher can run that calculation at scale, in plain language, without touching a line of code,” said Aayush Singh, Head of Science, Catalysis, SandboxAQ.

Marketing Technology News: How MarTech Is Enabling Autonomous Brand Engagement Across Channels?

The applications reach across green hydrogen, sustainable aviation fuel, fertilizer production, and plastics recycling, letting research teams pre-screen large design spaces before they commit DFT or experimental budgets and shortening innovation cycles that once took years.

Geoff Ling, Founding Director of the Biotech Office at DARPA, said, “In science, the biggest risk is often the unknown. Anything that helps researchers reduce uncertainty is incredibly valuable, and SandboxAQ’s MCP tools in Claude do that for both materials science and drug discovery.”