Quantifying and clarifying the spatial differentiation of ecosystem services demand from the perspective of micro-land use subjects is of great significance for improving the human well-being of farmers in resource-dependent areas. Taking Mizhi County in the loess hilly and gully region as a study area, we combined public participatory GIS and MaxEnt model to achieve spatial assessment of 10 ecosystem services demand. At the same time, K-means clustering was used to identify village ecosystem services demand bundles, and the impact of environmental variables on ecosystem services demand and its bundles spatial distribution was explored. The results were as follows:① The spatial differentiation of the ecosystem services demand was varied. In the provision service, grain supply and meat supply both showed a spatial differentiation of high in the middle and low in the east, and the spatial differentiation of fruit supply was the opposite. In the regulating service, the areas with high demand for soil conservation and air purification were distributed in high-altitude areas, and the areas with high demand for water conservation were distributed in the Wuding River Basin. Four types of cultural services all showed a spatial pattern of alternating distribution of high, medium and low demand areas, but the area proportions of demand areas at different levels varied significantly. ② Three types of ecosystem services demand bundles were identified using K-means clustering:including food supply demand bundle, the cultural entertainment demand bundle, and the ecological conservation demand bundle. Among them, food supply demand bundle was the dominant local ecosystem services demand bundles, mainly distributed in the north and south of the study area. The cultural entertainment demand bundle was mainly concentrated on both sides of the Wuding River. The ecological conservation demand bundle was in the south of the study area, next to the cultural entertainment demand bundle, and in the north was more dispersed. ③ The dominant environmental factors of the spatial differentiation of various ecosystem services demand were different. The environmental variables that affect the food supply demand bundle were ranked as:land use types> altitude> distance to the water body, the environmental variables that affect the cultural and entertainment demand bundle were ranked as:distance to the accommodation> distance to the water body> distance to the scenic spot, and the environmental variables that affect the ecological conservation demand bundle were ranked as:land use types>distance to the accommodation> distance to the water body. The research results revealed the spatial differentiation law of farmers' ecosystem services demand and its bundles in resource-dependent regions, which can provide references for ecosystem management and policy formulation in the region.