Abstract: A dynamic equivalent regulatory factor model was constructed to estimate the value of agroecosystem services in 31 provinces of China from 2005 to 2022. The Dagum Gini coefficient, kernel density estimation, Markov chain and geodetector methods were comprehensively used to explore the spatio-temporal evolution of agroecosystem services value and the associated driving factors. The results showed that: (1) China's agro-ecosystem services value generally exhibited a fluctuating upward trend from 2005 to 2022. In terms of the spatial distribution, China's agro-ecosystem services value displayed a ladder-type characteristic of “high in the south and low in the north, high in the east and low in the west”. (2) The overall difference in the agro-ecosystem services demonstrated a U-shaped evolutionary trend of fluctuation, with hypervariable density being the primary source of the overall difference, followed by inter-regional differences, and intra-regional differences having the lowest contribution rate. During the study period, regional imbalance increased significantly, polarization trends were evident, and internal mobility remained weak. (3) Natural geographic factors drove the spatio-temporal evolution of the agroecosystem services more strongly than the economic and social factors. Precipitation was the dominant factor that influenced the spatial and temporal evolution of the agroecosystem services, and the interactive combination of precipitation and GDP per capita was the key driver.