Granular column collapse (GCC) is a representative large-deformation benchmark problem that enables systematic investigation of granular failure, post-failure flow, and deposition processes under well-defined boundary conditions [1]. In this study, GCC was simulated using the Material Point Method (MPM) coupled with a state-dependent constitutive model formulated within the framework of critical state soil mechanics, incorporating material state dependency [2], plastic strain evolution, strain hardening and softening responses, and non-associative flow (f ≠ g). A nonlinear isotropic elastic formulation based on energy conservation principles was adopted to incorporate the effects of mean effective and deviatoric stresses [3]. The proposed numerical framework was first benchmarked against the experimental observations of Lube et al. [4], demonstrating close agreement in runout distance and final deposit profile. Subsequently, a systematic parametric study was performed to investigate the influence of basal friction and gravity conditions on collapse behaviour. The results indicated that basal friction played a significant role in controlling the final deposit height and width. In contrast, gravity conditions had a limited influence on deposit geometry but governed the potential energy of the mobilised mass. Furthermore, the initiation of failure was delayed with decreasing gravity. These findings provided improved insights into the mechanics of granular collapse and demonstrated the proposed MPM framework’s capability for large-deformation geotechnical problems.