Cavitation typically occurs when the fluid pressure is lower than the vapor pressure at a local thermodynamic state. The goal of our overall efforts is to establish a predictive tool for turbulent cavitating flows, including those under cryogenic conditions with noticeable thermal effect associated with the phase change. The modeling framework consists of a transport-based cavitation model with ensemble-averaged fluid dynamics equations and turbulence closures. To date, the reported experimental investigations contain little information about the turbulent characteristics in the flow field. However, we have observed that inlet turbulent quantities can substantially impact the outcomes for cavitating flow. Because the exchange between static and dynamic pressures has a dominant impact on the cavitation dynamics, and the viscous effect can modify the effective shape of a solid object to cause noticeable variations in the predicted multiphase flow structures. A filter-based approach is utilized along with two-equation turbulence closures so that one can assess the local numerical resolution with the computed turbulence length scale, and reduce the impact of the inlet boundary conditions of the eddy viscosity. The effectiveness of the simulation framework is confirmed by using experimental data covering both isothermal and cryogenic cavitation with different geometries.