Although climate conditions primarily determine the distribution and functioning of vegetation, vegetation also influences climate via biophysical and biogeochemical features such as evapotranspiration, albedo, carbon cycling, trace gas emissions and the roughness of the land surface. Forecasts of rapid climate change during the next 100~200 years, fueled by an increase in greenhouse gases, have motivated the development of land surface models (LSMs) that predict changes in vegetation functions. Here, we review how these models have been developed and used to simulate interactive processes between climate and the land surface. Current limitations and future perspectives of the LSMs are also presented.