In this study, simulations using both the finite-rate chemistry (FRC)-LES and the flamelet/progress-variable (FPV)-LES approaches are conducted for a piloted partially premixed methane/air flame with high turbulence intensity. The two models have different spatial distributions of both time-averaged quantities and instantaneous flame field. For both axial and radial profiles of time-averaged statistics, the FPV-LES approach provides overall better prediction than FRC-LES, primarily due to the unity effective Lewis number under high turbulence intensity. To properly apply FRC in LES, a better transport model covering a broad range of turbulence intensity is required. In contrast, for conditional statistics, in which the effects of transport modeling are largely removed, the FRC-LES approach provides overall better predictions than FPV-LES for all quantities at most locations and mixture fractions.