Rapid urbanization has led to economic growth with inevitable air pollution. There are significant differences in the dominant factors of different air pollutants. However, the influencing mechanism of urbanization on different air pollutants is still unclear. Therefore, exploring the differential effects of urbanization on various air pollutants is of great significance for accelerating local collaborative treatment of air pollutants and improving regional air quality. Based on the analysis of the spatial–temporal pattern evolution, spatial agglomeration, and internal correlation of six air pollutants in mainland China during 2013–2020, namely PM2.5, PM10, SO2, NO2, O3 and CO, we combined the environmental Kuznets theory to build a panel regression model of urbanization on six air pollutants. Except for NO2 and O3, the concentrations of the other four air pollutants all decreased to different degrees, among which, SO2 concentration decreased the most. The spatial pattern of air pollution showed that the concentration of air pollutants in typical areas decreased significantly, while which in areas with higher population density or higher economic development was relatively higher. As a key factor affecting air quality, different aspects of urbanization have significant differences in the direction and intensity of effect on various air pollutants. The relationship between the concentration of six air pollutants and urbanization conforms to the differential environmental Kuznets curve (EKC). There are nonlinear relationships between urbanization and PM2.5, PM10, SO2, NO2, O3 and CO concentrations, which are inverted “U-shaped”, inverted “U-shaped”, inverted “U-shaped”, inverted “N-shaped”, “U-shaped”, and inverted “U-shaped” respectively. In addition, urbanization has a spillover effect on PM2.5, PM10, SO2 and CO concentrations, while the direction of whose spillover effect reflects the phased change.