EXTENDED ABSTRACT: The thermodynamic oxide database plays a foundational role in numerous scientific and engineering ffelds, such as metallurgy, materials science, and chemical engineering, providing critical data for understanding and optimizing oxide systems. The recent advancements in the development of the GTOx oxide database will be presented, with a focus on the unification of the thermodynamic description of the liquid state and the inclusion of thermophysical properties such as viscosity and density of oxide melts into the database. In addition, case studies will illustrate how the thermodynamic oxide database supports the development of advanced materials with tailored properties, such as hightemperature stability and corrosion resistance, and assists in the prediction and control of complex equilibrium in metallurgical and chemical processes. Combination of thermodynamic database and machine learning in predicting slag properties will be also given as a demonstration, which will accelerate the study on innovative combination of the data-driven model with the mechanism model. It aims to highlight the database’s growing relevance and adaptability in addressing contemporary challenges in materials research and industrial applications, showcasing its potential to drive future innovations.

Keywords：Thermodynamic database; Oxide systems; Modelling; Industrial application; Machine learning