Global evapotranspiration (ET) products, as compensation for eddy-covariance observations, provide useful data sources for understanding terrestrial water-energy budgets at different scales, especially for data-sparse regions. Here, we evaluated three remotely sensed ET products against water balance-based reference ET () in 16 river basins across the Tibetan Plateau (TP) on a monthly time scale from 1983 to 2011. The results indicated that ET_GLEAM performed the best overall across the 16 TP river basins in terms of the multi-year average and the interannual variability of monthly , followed by ET_ZHANG and ET_CSIRO. The multi-year means of monthly were better estimated overall by the three remotely sensed ET products rather than their interannual variability. However, the performances of the three ET datasets varied among different TP basins based on various evaluation criteria. The seasonal cycle of was better captured by ET_GLEAM, ET_ZHANG and ET_CSIRO in the Yalong, Yangtze and Salween Basins and the upper Yellow River Basins rather than that in the Yulongkashi, Bayin and Brahmaputra River Basins. Overall, the ET_GLEAM performed relatively better than other datasets. The evaluation results will provide important references for us to select suitable datasets and to apply them in basin-scale water-energy budget studies in data-sparse regions.