Regional landslide hazard assessment based on physical-mechanical models has currently become a major research issue for landslide risk prevention. However, the accurate and automatic determination of key parameters of the models remains a challenge. Most of the existing parameter determination methods are highly subjective and time-consuming. This study introduces an innovative framework for quantitative landslide hazard assessment in the Longyang to Yanguo Gorge section of the upper Yellow River, China. It integrates Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) technology with the Scoops3D slope stability model. SBAS-InSAR detects slow-deforming slopes, which act as a calibration base for automating the inversion of geotechnical parameters, such as cohesion and internal friction angle, in the Scoops3D model, while addressing spatial heterogeneity through distinct rock group divisions. A Python-based automated inversion system, utilizing confusion matrix evaluation, is developed to calibrate parameters across geological units. The calibrated Scoops3D model is used to assess the landslide hazards under natural and seismic conditions. The results show that geotechnical parameters inverted from SBAS-InSAR deformation data are generally higher than those inverted from historical landslides. The multiple-parameter model based on InSAR data achieves the highest predictive accuracy, with an area under the ROC curve (AUC) of 0.85, outperforming both the single-parameter InSAR model (AUC = 0.82) and the historical landslide-based model (AUC = 0.73). These findings demonstrate the enhanced reliability and practicality of InSAR-informed models for landslide risk assessment.