The regional Climate-Weather Research and Forecasting model (CWRF) was used to downscale the National Climate Center operational short-term climate prediction based on the Beijing Climate Center Climate System Model (BCC_CSM1.1m). The prediction skill was assessed for daily precipitation and surface air (2-m) temperature in March-August during 1991-2010. As compared with BCC_CSM, CWRF predicts more realistic seasonal mean precipitation and temperature spatial distributions and resolves more detailed features over the mountainous regions with large terrain variations. It also predicts more accurately precipitation interannual variations in East and Central China. Overall, CWRF outperforms BCC_CSM in predicting precipitation of different intensities, especially extreme events. Benefited from higher spatial resolution and advanced physical process simulation at lower levels, CWRF downscaling can improve seasonal prediction of summer precipitation in China.