On June 24, 2017, a catastrophic rock avalanche with a volume of 5.08 × 10⁶ m occurred in Xinmo village, Maoxian County, Sichuan Province, SW China. Several studies focusing on the triggering factors, failure mechanism and dynamic process of the Xinmo landslide have been reported in the literature. However, there are few studies with detailed three-dimensional investigations based on numerical modeling that assess the post-failure stage of the landslide. In the present study, the three-dimensional material point method is used to reproduce the kinematics of the landslide to obtain better insight into the disaster by considering the effects of fragmentation and entrainment. Based on this framework, a velocity-weakening friction law is implemented to describe the basal friction between the moving mass and the substrate, and a strain–softening Drucker–Prager constitutive model is used to describe the internal deformation of the moving mass. The entrainment processes obtained with and without the proposed model are compared, and the simulated results generated with the presented models showing the distribution characteristics of the materials in the deposit area match the observed results fairly well. The implementation of a velocity-weakening friction law and a strain-softening Drucker–Prager constitutive model in the material point method may provide valuable insight for the enforcement of physically based risk assessments of “rockslide-debris avalanches”.