Abstract: The stability of the heterogeneous traffic flow mixed by two driving styles was studied. The corresponding car-following models of the two driving styles, the optimal velocity car-following model and the intelligent driver car-following model, were calibrated first using the real traffic data. The stability of the heterogeneous traffic flow was derived, based which the stability functions of the traffic flow and the driving style were defined. The simulations were conducted to analyze the stability condition. The results show that the stability of the heterogeneous traffic flow is impacted by the two factors: the proportion of the driving style and the stability function of the driving style. The simulation results showed that the intelligent driver driving style always has better stability effect than the optimal velocity driving style. In addition, the neutral stability lines with respect to the proportions of the two driving styles are obtained under the rage of the whole feasible velocity. The results provide the theoretical supports for the control of the heterogeneous traffic flow stability and the traffic congestion.