Suitable air distributions are essential for creating thermally comfortable and healthy conditions in indoor spaces. Computational fluid dynamics (CFD) is widely used to predict air distributions. This study systematically assessed the performance of the two most popular CFD programs, STAR-CCM+ and ANSYS Fluent, in predicting air distributions. The assessment used the same meshes and thermo-fluid boundary conditions for several types of airflow found in indoor spaces, and experimental data from the literature. The programs were compared in terms of grid-independent solutions; turbulent viscosity calculations; heat transfer coefficients as determined by wal functions; and complex flow with complicated boundary conditions. The two programs produced almost the same results with similar computing effort, although ANSYS Fluent seemed slightly better in some aspects.