In recent years, frequent heat waves have severely reduced the quality of human settlements in big cities. To reveal the influence of urban spatial form on land surface temperature (LST) in mountainous cities, Chongqing city in China was selected as a case study. The quantitative relationships between five urban spatial form factors and LST are analyzed, such as sky view factor (SVF), urban surface roughness (USR), road surface area density (RSD), building density (BD) and floor area ratio (FAR). The results showed that the spatial scale influence their correlations. Most urban spatial form factors have a significant positive relationship with LST, besides FAR. The various regression results have also shown that spatial regression models (Spatial lag model, Spatial error model and General spatial model) are working better than OLS model, due to the reason that some of the urban spatial form factors have spatial spillover and spatial heterogeneity effect. Besides the spatial effect, the unique topography of mountainous city also has an impact on LST, such as the slope, aspect and underlying surface. Therefore, the planners should consider those differences in mountainous cities, and using different methods to mitigate UHI and making the city cope with climate change.