The currently available computational power limits the utilization of direct numerical simulation (DNS) in practical engineering flow applications. In recent years, large-eddy simulation (LES) has emerged as a viable high-fidelity tool for such flow problems; however, it suffers from the same computational limitations in the near-wall region. In this talk, I will introduce a new way of modeling the wall in LES to overcome the limitation of the near-wall region. First, the use of the slip boundary condition with transpiration for wall-modeled LES is motivated by theoretic assessment and a priori testing using DNS data. Secondly, a dynamic slip wall model consistent with the constant stress layer in the near-wall region is presented. The dynamic slip wall model is free of any a priori specified coefficients, unlike traditional wall models which are based on RANS models. The results show that the predictions of the mean velocity profile and turbulence intensities from the dynamic slip wall model are in good agreement with DNS and experimental data for a wide range of Reynolds numbers and grid resolutions for canonical turbulent boundary layers.