Soil moisture is an important parameter in the process of land-atmosphere interactions research. It plays a significant role in atmospheric circulation, climate change and drought research. Lacking of soil moisture observation, reanalysis of soil moisture data plays an important role in operation and research. Evaluation and comparison of soil moisture (0-10cm) from May to October in 2018 over Inner Mongolia by using the Soil Moisture Operational Products System data, Global Land Data Assimilation System version2.1 data (NOAH) , European Centre for Medium-Range Weather Forecasts Reanalysis Interim data and European Centre for Medium-Range Weather Forecasts fifth Reanalysis data was made. After the analysis of temporal and spatial variation of soil moisture, results show that: SMOPS, GLDAS2.1, ERA-Interim and ERA5 soil moisture could be used to reproduce the spatial and temporal distribution with over estimation. The ERA5 and GLDAS2.1 are better than ERA-Interim and SMOPS with over estimation in 3 areas over Inner Mongolia. ERA5 has the highest correlation coefficient with higher Root Mean Square Errors. And reanalysis soil moisture is better than soil moisture retrievals from satellites/sensors.