Abstract: Under the background of the ‘Dual Carbon’ goal, understanding the development trend and driving mechanism of agricultural carbon emissions in the Yellow River Basin is of great significance for exploring agricultural carbon emission reduction paths and achieving low-carbon agricultural development. Based on the calculation of agricultural carbon emissions of 56 prefecture level and above cities in the Yellow River Basin from 2011 to 2021, this study uses Kernel Density Estimation, Dagum Gini coefficient, and Spatial Durbin Model to examine the spatiotemporal evolution, spatial differences, influencing factors, and spatial effects. The results indicate that: (1) From 2011 to 2021, agricultural carbon emissions in the Yellow River Basin decreased from 121.15×10⁶ t to 108.23×10⁶ t, going through three phases: ‘slow increase-continuous decrease-slight rebound’. The evolution process and polarization characteristics of agricultural carbon emissions vary across the upstream, midstream, and downstream areas. (2) Agricultural carbon emissions in the Yellow River Basin show significant spatial agglomeration and regional differentiation. The upstream areas exhibit a ‘high in the north, low in the south’ spatial pattern, the midstream areas show a ‘high in the west, low in the east’ distribution, and the downstream areas feature a ‘center-periphery’ spatial structure. (3) Overall differences in agriculture carbon emissions in the Yellow River Basin, as well as intra-regional differences in the midstream and downstream areas show a fluctuating downward trend. Inter-regional differences are the primary source of overall differences, with an average annual contribution rate of 43.87%. The difference between the midstream and downstream areas is the largest, while the difference between the upstream and midstream areas is the smallest. (4) Government intervention and the intensity of agricultural mechanization suppress agricultural carbon emissions, and government intervention shows a significant negative spatial spillover effect. The per capita cultivated land area and the agricultural industrial structure dominated by planting and animal husbandry have a positive impact on agricultural carbon emissions. The impact of agricultural industrial agglomeration and agricultural economic development level on agricultural carbon emissions in local and surrounding areas follows an inverted ‘U’ shape. The study reveals that there are significant spatial and temporal differences in the evolution characteristics and influencing factors of agricultural carbon emissions in various regions of the Yellow River Basin. Differentiated regional carbon emission reduction strategies should be formulated and implemented to strengthen regional exchanges and cooperation on emission reduction technologies and experience sharing, thereby promoting low-carbon and sustainable development of agriculture.