Shape is one of the most important properties of particles, and can affect particle flow behavior significantly in particulate systems. In the past, extensive studies have been conducted on the effect of particle size and density on the mixing quality of particle mixtures in gas-fluidized beds, but little is known regarding the influence of particle shape. In this work, CFD-DEM approach is used to investigate the mixing of binary mixtures composed of ellipsoids and spheres. A modified drag model suitable for multicomponent mixtures of non-spherical particles is proposed first, and its validity is then verified by comparison to experimental data. The simulation results show that for the cases considered, adding a second component of ellipsoids to spheres results in reducing the minimum fluidization velocity of the mixtures; however, oblate particles in decreasing the minimum fluidization velocity is more significant than prolate particles. The mixing index of all binary mixtures generally increases with increasing gas superficial velocity. The results also show that with the added component varying from spheres to oblate or prolate particles, segregation happens in the bed and becomes more severe with aspect ratio. It is found that the effect of particle shape on the drag force is responsible for the occurrence of particle segregation.