Jua

Jua is developing a foundation model that accurately represents how reality behaves and includes an agent that operates within it. This innovative model primarily focuses on atmospheric prediction, demonstrating state-of-the-art performance compared to traditional systems. By understanding the complexities of the atmosphere, Jua can extend its capabilities to various other domains governed by physical laws.

With its advanced technology, Jua aims to revolutionize the way industries approach challenges related to energy, commodities, and physical systems. The core components include a world model, EPT-2, which learns physics from extensive datasets, and an agent named Athena that executes objectives in physical contexts. This dual approach not only provides predictive analytics but also enables proactive decision-making within utilities and trading environments.

As Jua's foundational model gains traction, it is being adopted by major companies in the energy sector, including TotalEnergies, Shell, and Hydro-Québec, showcasing its practical utility across applications. The platform’s development is driven by the belief that better models for the physical world can lead to better decisions and innovations across various industries.