Grinding is considered one of the most complex manufacturing processes due to the stochastic nature of wheel topography and the interactions between abrasive grains and the workpiece. Therefore, accurate simulation plays a critical role in defining uncut chip thickness and contact length, as well as predicting surface quality and optimizing process parameters. In this study, a comparative analysis is presented between two-dimensional (2D) and three-dimensional (3D) grinding simulation frameworks based on the Z-map technique. The models were developed using data from a foundational study on grinding wheel topography, enabling a consistent evaluation of both approaches in terms of computational efficiency and accuracy. The results show that 2D models can provide a high level of accuracy in simulations with constant parameters, with deviations not exceeding 0.25% compared to 3D results. However, in stochastic simulations, the 2D approach tends to underestimate key parameters such as uncut chip thickness (by up to 60.3%) and contact length (by approximately 19.7%). This limitation is mainly attributed to the inability of 2D models to capture off-center grain interactions. On the other hand, despite their high computational cost—requiring up to 4–5 days on a single-core system—3D simulations provide a more realistic representation of the grinding process. The results show strong agreement with experimental measurements of surface roughness (Ra = 2.1 µm compared to 1.9 µm experimentally). Recent developments between 2016 and 2026, particularly in parallel computing and active-point selection techniques, have significantly improved the efficiency of 3D simulations. These advancements help reduce the gap between the efficiency of 2D models and the accuracy of 3D approaches, paving the way toward real-time, high-fidelity kinematic simulations. To overcome the limitations of full 3D modeling, a hybrid 2D–3D simulation framework is proposed. This approach combines the efficiency of 2D models with the accuracy of 3D simulations, reducing computational time from several days to a few hours while maintaining a high level of predictive accuracy............... Keywords: .........Grinding simulation, Hybrid 2D–3D modeling, Z-map technique, Uncut chip thickness, Surface roughness, Grinding wheel topography.