Agricultural productive services are fundamental in transforming the paradigm of agricultural development and exert a significant impact on farmers’ behavior in managing their arable land. This paper employs the slack-based measure of the super-efficiency model, known as the Super-SBM model, incorporating unexpected output to gauge the eco-efficiency of cultivated land use (ECLU). Utilizing panel data from Hunan Province spanning the years 2007 to 2020, this study unveils the following key findings: (1) Across various quantile levels, agricultural productive services exhibit a substantial capacity to enhance ECLU, with coefficients ranging from 0.070 to 0.156. (2) The impact of agricultural productive services on ECLU is constrained by both farmers’ income levels and the scale of cultivated land, revealing a dual threshold effect. As income levels rise, the corresponding coefficient increases progressively, from 0.0771 to 0.1147, and ultimately to 0.1571. Similarly, with the expansion of cultivated land, the coefficient increases from 0.1152 to 0.1443, and ultimately peaks at 0.1694. (3) The enhancement of ECLU through agricultural productive services is achieved by reducing the input of environmental factors and mitigating undesirable outputs. (4) Agricultural productive services indirectly facilitate labor transfer, accounting for 10.8% of the overall effect. Consequently, it is imperative to continually enhance the agricultural socialized service system, bolster financial support, and implement policies that foster the development of agricultural productive services to realize sustainable cultivated land utilization.