Creating a world where smaller teams can engineer and manufacture new advanced materials in months, not decades.
PHIN is building the world's highest performing foundation models for matter simulation, which evaluates material properties at an unprecedented scale, speed, and accuracy. Our proprietary platform empowers engineers to build and run digital experiments easily, allowing them to answer bespoke materials questions with confidence—all without the need for expensive and time-consuming lab work.
Foundation Models
PHIN-atomic enables prediction of materials properties from first-principles through a novel neural network architecture that delivers quantum-accurate simulations. Efficient active learning ensures models are continuously learning, generating trustworthy, scalable predictions across complex materials.
PhinOS, our proprietary platform, gives materials engineers access to artificial intelligence tools to accelerate the entire materials development process – creating atomic blueprints, predicting materials performance, analyzing results, and proposing new materials candidates.
Our foundation models are designed to be general-purpose, allowing us to simulate a wide range of materials and their properties. This means we can address a variety of challenges across different industries, from energy storage to carbon capture, and beyond.
Batteries
We empower battery companies to improve key battery performance areas, such as energy density, cycle life, and safety.
