Abstract: To address the problem of low evaluation accuracy of the Sinmap model in evaluating the stability of shallow loess landslides under the action of rainfall, a method based on the Maxent-Sinmap model is constructed to evaluate the stability of regional shallow rainfall loess landslides under the action of rainfall by improving the evaluation of the Sinmap model based on the maximum entropy model.Taking Zhidan County, Shaanxi Province, a high-incidence area of loess landslides, as an example, the relevant data of topography, geotechnical parameters and geological disasters were obtained by field and indoor work. The main environmental variables were obtained by Maxent model, and then the main environmental variables were partitioned. The Sinmap model was used to evaluate the stability of shallow loess landslides in different partitions under rainfall. The results show that based on the Maxent model, the landslide in Zhidan County is mainly affected by five indicators, such as slope, rainfall, landform, road buffer zone and normalized vegetation coverage index. The contribution rates to historical disaster points are 27.1 %, 20.3 %, 18.8 %, 18.7 % and 6.2 %, respectively. Compared with the traditional Sinmap model, the density of disaster points in the unstable area of the model increased by 17.26 %, 16.54 %, 17.39 %, 14.20 % and 12.96 % respectively under the conditions of light rainfall, moderate rainfall, heavy rainfall, rainstorm and downpour. The results of the Maxent-Sinmap model have a larger stable area than those of the Sinmap model, and there is no historical disaster distribution in the expanded area of the stable area. The model has higher accuracy and more reliable results, which can provide a better scientific basis for the evaluation of regional shallow rainfall landslides.