Google Cloud adds SandboxAQ AI models to marketplace for scientific calculations

Google has connected Large Quantitative Models (LQM) from SandboxAQ to its cloud marketplace — AI trained not on internet texts, but on scientific equations and experimental data. The new models will work alongside Gemini and be available through the standard Google Cloud Marketplace.

Why LLMs Fall Short in Science

Large language models can do many things: summarize articles, write code, explain concepts in plain language. But they have a systematic problem with mathematics and precise calculations. Models often make errors in computations, confuse physical constants, and cannot work with equations the way real researchers need them to. The reason is architectural: LLMs are trained to predict the next token in a sequence of text. For them, numbers are symbols, not quantities. A model doesn't "understand" that 0.001 moles of substance and 1 mole are different orders of magnitude. For tasks in chemistry, physics, or biology, such an error can mean experimental failure.

"Language models look convincing when they write about science, but

that's not the same as calculating correctly."

What LQM from SandboxAQ Is

SandboxAQ is a company that spun out from Alphabet (Google's parent company) in 2022. Its focus is quantum simulations and AI for scientific tasks. Large Quantitative Models are a fundamentally different class of AI: instead of predicting text, the models are trained to work directly with mathematical structures and numerical data. LQM is based on training with structured scientific datasets:

• Equations from physics, chemistry, materials science, and biology
• Laboratory data and experimental results
• Numerical methods and algorithms for simulation
• Parameters of molecular structures and material properties

The result is models that solve computational problems with precision unattainable for a general-purpose LLM.

How the LQM and Gemini Integration Works

The LQM + Gemini combination opens fundamentally new scenarios for corporate clients to use cloud AI. Previously, researchers had to choose: either work with precise but narrowly specialized scientific tools, or use general-purpose LLMs that accelerate text work but are unreliable in calculations. Now the roles are clearly divided: Gemini handles textual context, formulates the problem, and explains results, while LQM handles precise computations. Potential areas of application include development of new drug compounds, climate modeling, creation of materials for electronics, and quantum simulations. Access through Google Cloud Marketplace means researchers can use LQM without needing to deploy their own infrastructure.

What This Means

The inclusion of specialized "scientific" AI in Google Cloud is an acknowledgment that one universal model doesn't solve all problems. The era of "one LLM for everything" is gradually giving way to a mosaic of specialized tools. For research organizations, this opens the possibility to implement AI in laboratory processes without compromising between speed and accuracy. Google, for its part, is expanding Cloud beyond corporate automation — toward science and deep tech.