Abstract: Landslide hazard assessment is a crucial component of regional landslide disaster risk warning and control. The distributed slope stability quantitative evaluation model SINMAP (Stability Index MAPping) is widely used in landslide hazard assessment because it effectively mirrors the physical and mechanical mechanisms underlying slope stability. However, traditional SINMAP models overlook the spatial differences in rock and soil characteristics due to geological environmental changes, resulting in low accuracy in assessment results. To address these deficiencies, this paper explored an enhanced SINMAP model tailored to various spatial calibration zones.Using Dazhou Town, Wanzhou District, Chongqing as a case study and employing frequency ratio and sensitivity index analysis, the key indicator factors are determined as rock and soil type, vegetation coverage, and proximity to roads from the eight indicator factors reflecting the cause of landslides. Based on the spatial distribution differences of key indicators, the study area was segmented into 6 different spatial calibration zones, and each assigned corresponding physical and mechanical parameters of the rock and soil. A comparative study of landslide hazard assessment using both traditional SINMAP and its improved models was conducted. The results indicate that: 1) Overall, the high and extremely high-risk landslide zones predicted by the two models are mainly distributed along the reservoir bank, adjacent to the river, and areas with strong engineering activities in the study area. 2) Under the most dangerous working conditions, the improved SINMAP model achieved an AUC value of 86.8%, surpassing the traditional model’s AUC of 73.9%, and enhancing the accuracy of landslide recognition by 12.9%. 3) In the local calculation results of landslide disasters, 81.82% of the actual landslide points were in areas above the medium risk level under the most dangerous working conditions, compared to 72.73% in the traditional SINMAP model. Therefore, the improved SINMAP model offers superior detection capabilities, a more continuous spatial distribution of detection results, and more accurate alignment with the real-world characteristics of landslide development.