Metal additive manufacturing (AM) processes regularly experience high temperatures, quick velocities and rapid solidification taking place in a small volume melt pool. Accurate capturing of data from the melt pools is very difficult and can require expensive equipment. Physics based numerical models can be used to simulate the AM process. However, there are many thermal and fluid dynamics models that have been developed each with their own sets of physical models and assumptions. In this work, a three dimensional multiphysics computational fluid dynamics model was used to obtain thermal behavior, geometry and surface topology from the melt pool of single laser scan tracks on a bare Inconel 625 substrate. The simulation results were compared with the highly controlled AMB2018-02 series of benchmark tests performed by the National Institute of Standards and Technology (NIST). The complex physics involved in the simulations, including heat and mass transfer, fluid flow, evaporation, and Marangoni convection enabled accurate prediction of the melt pool geometry, thermal behavior and the surface topology. The simulation results for the thermal behavior were mostly within the standard deviation of the experimental results. Simulation results from surface topography were able to capture the peak and valley trends. Overall, the underlying physics involved produced good results in a reasonable amount of time and were accurate enough to form a basis that can be coupled with other models to predict microstructure and other material properties.