Landslides are a serious threat to infrastructure and human settlement in mountainous regions, particularly in the upper Teesta River basin of the Darjeeling-Sikkim Himalaya. This chapter employs three popular techniques—logistic regression (LR), frequency ratio (FR), and analytical hierarchy process (AHP)—to map the frequency of landslides. The objective of this chapter is to quantify the effectiveness of these models and offer practical insights into their performance in landslide risk assessment. Using data from remote sensing and field investigations, this chapter provides a comprehensive inventory of landslides. Numerous theme layers, such as slope, aspect, elevation, plan curvature, profile curvature, roughness, lithology, lineament density, land use land cover, NDVI, road density, drainage density, annual rainfall, TWI, and SPI, are thought to be likely causes of landslides. The predictive power of each model is assessed using the AUC of ROC curves. When evaluating landslide susceptibility, the ROC curve results demonstrate that all three models perform well, with AUC values better than 0.70. Among the models, the Logistic Regression (LR) model performs the best, with the highest AUC value of 0.982. The FR model and the AHP model trail closely, with AUC values of 0.951 and 0.703, respectively. By using these models, a more accurate and long-lasting landslide susceptibility map is produced, which offers vital information for resource distribution, disaster prevention, and land-use planning in the Darjeeling-Sikkim Himalaya. It contributes to the ongoing efforts to enhance landslide risk assessment approaches and encourages the development of effective strategies for managing landslide risks in mountainous environments.