The growth and evolution of biofilms in porous media involve complex coupled physicochemical and biological processes. Their pronounced multi-scale characteristics, heterogeneity of the media, and uncertainties in model parameterization have led to fundamental divergences in the theoretical frameworks of numerical models across different scales. This poses significant challenges for the accurate characterization and prediction of biofilm dynamics. In recent years, advances in computational techniques have driven substantial progress in pore-scale, continuum-scale, and cross-scale coupled numerical modeling and simulation of biofilm growth. However, considerable bottlenecks remain in model development, validation, and utilization. These include difficulties in characterizing three-dimensional microscopic pore structures, the complexity of constructing biofilm growth dynamics models, the lack of quantitative standards for cross-scale multiprocess coupling strategies, and the scarcity of experimental data required for model parameterization. Based on the mechanisms of biofilm growth dynamics in porous media, this paper reviews the research progress of pore-scale, continuum-scale, and multi-scale coupling numerical models, analyzes the theoretical foundations, numerical algorithms, application cases, applicability, and limitations of biofilm growth models at different scales. It also summarizes the application potential of three-dimensional imaging technologies, outlines the emerging trends in mechanistic representation of the complete biofilm growth processes, and explores the optimization pathways for cross-scale coupling modeling strategies. This review provides a theoretical basis for the selection and improvement of biofilm growth models, and offers technical support for the engineering application of soil microbial technologies in environmental pollution control and ecological restoration.