Urbanization is a key factor that threatens the stability of ecosystem services (ESs), which are crucial for maintaining ecological security and enhancing human quality of life. Gaining insight into the spatiotemporal differentiation of the ecosystem service value (ESV) and its coupling relationship with urbanization is a key issue in promoting sustainable regional development. We employed various algorithms, including the improved equivalence factor method, random forest model, mixed geographically and temporally weighted regression model, and coupling coordination degree (CCD) model, to reveal the spatiotemporal evolution of the ESV and its driving mechanisms in the Lanzhou-Xining urban agglomeration (LXUA) from 1980 to 2020. In addition, we explored the coupling relationship between ESV and the combined urbanization index (CUI). The results showed the following: (1) From 1980 to 2020, the interannual variation in ESV first decreased and then increased, with an increase of 230 million yuan/annum in 2020 compared to that of 1980. Spatially, the ESV exhibits a distribution pattern of high in the south and low in the north, and high in the west and low in the east. (2) Land use intensity (LUI) and elevation contributed degree values exceeding 20% to the ESV and were the most important drivers of the ESV. (3) From 1980 to 2020, the average CCD values of the ESV and CUI transitioned from being severely unbalanced (0.19) to being slightly balanced (0.42). There was a negative spatial correlation between the ESV and CUI, with each cluster type was distributed discretely in space. Our study emphasizes that in areas characterized by robust ecosystem integrity and a high ESV, the increase in LUI was constitutes the main reason for the decline in the ESV. The findings of this study can provide a scientific basis for the coordinated development of ESVs and urbanization in the LXUA and other cities.