Various drag models were implemented in a superquadric CFD-DEM code and validated for the simulation of spherical and ellipsoidal particles in spouted beds. The most suitable drag models were identified by comparing the predicted local particle velocity with those measured by particle tracking velocimetry (PTV). The model by Beetstra et al. is the one that best reproduces the experimental results for the spouting of spherical particles, whereas the one by Sanjeevi et al. is the most suitable for ellipsoidal irregular ones. Their capability for the prediction of key operating parameters was demonstrated in both conical and prismatic spouted beds, as they correctly predict the minimum spouting velocity (ums) and fountain height in different configurations (without draft tube and different draft tubes) for particles of different size and shape. The CFD-DEM model predicts the preferential orientation of ellipsoidal particles at each location in the bed and the influence of internal devices on this parameter, which is of the utmost importance in the design of reliable coating apparatuses in the pharmaceutical industry.