The geographical environment in western Sichuan is complex, and the spatial structure of landslides is diverse and the deformation is severe. In order to effectively prevent landslide disasters, it is necessary to explore the reliable evaluation method of landslide hazard in related areas. The G1-entropy weight coupling weighting TOPSIS (technique for order preference by similarity to ideal solution) was used to construct the landslide hazard assessment model for the first time, and 12 evaluation factors such as elevation, rainfall and ground motion parameters were selected to evaluate the landslide hazard of the Yalong River. According to the closeness degree, the landslide area was divided into four grades, namely, high-risk area (0.75 ~ 1), higher-risk area (0.5 ~ 0.75), medium-risk area (0.25 ~ 0.5) and low-risk area (0 ~ 0.25). At the same time, the measured results of four regional landslides obtained by SBAS-InSAR (small baseline subset interferometric synthetic aperture radar) technology and the analysis results of the evaluation model were compared. The G1-entropy weight coupling weighting TOPSIS method, the closeness degree of landslides in Jiulong County, Xinlong County, Dege County and Ganzi County are 0.854、0.686、0.405、0.224 respectively. The local deformation rates of landslide front edge obtained by SBAS-InSAR technology are -150、-43、-66、30 mm/a respectively. There is a significant correspondence between the two, which proves the accuracy of the evaluation model. The research methods and conclusions of this paper can provide theoretical basis and technical reference for the evaluation of landslide geological disasters in other regions of China.