This talk focuses on the application and capability of molecular dynamics (MD) and phase-field crystal (PFC) models in predicting qualitative and quantitative behavior of metals during solidification process. Both MD and PFC are nano-scale computational models. While MD is a well-established computational model, PFC is a more recently developed computational model that can act on diffusive time-scale necessary for the simulation of the solidification process. However, PFC models require further developments and also more efficient numerical algorithms for computational efficiency. By presenting numerous simulations of solidification in Fe, Ni, Cu, Al, and Mg, we show that a coupled MD-PFC approach is a promising computational tool to study this phenomenon in metals. The results of these simulations in predicting solidification, grain growth, and defects formation and also in predicting properties such as solid-liquid interface free energy, surface anisotropy, grain boundary free energy, and elastic constants will be also presented.