The paper titled Universal Physics Transformers presents a transformative learning paradigm able to model both Lagrangian and Eulerian discretization schemes. The approach, named Universal Physics Transformers (UPTs), is grid- and particle-agnostic, and it’s showcased across multiple fluid dynamics and particle dynamics applications. This innovative framework has the potential to significantly impact computational fluid dynamics and related fields, offering an adaptable and highly efficient model for various physical simulations.