Abstract: Green utilization of urban land has become the key to promote sustainable development in many countries and regions. Based on the relevant data of inputs and outputs of 76 prefecture-level cities in the Yellow River Basin from 2003 to 2021, this paper constructs an SBM model considering non-expected outputs to empirically measure the efficiency of green land utilization in urban land in the Yellow River Basin, and employs the standard deviation ellipse, Dagum Gini coefficient, and geographic detector to reveal its spatial and temporal patterns, sources of differences, and driving factors. It is found that: (1) the overall level of urban land green utilization efficiency in the Yellow River Basin is low, but it shows a slow growth trend during 2003-2021, with a significant increase in high-efficiency cities, and the center of gravity generally moves to the northwest; (2) inter-regional variations are the main source of the overall differences, and hypervariable densities and intra-regional variations make relatively small contributions; (3) the land type, technological level, population density are the dominant factors of land green utilization efficiency in the Yellow River Basin, and the interaction strengths of the driving factors are all greater than individual factors. The conclusions of this study are of great practical significance for promoting the continuous improvement of land green utilization efficiency in the Yellow River Basin and the sustainable development of the region.