Plant phenological events are sensitive indicators of climate change, and their change could markedly affect the structure and function of ecosystems. Previous studies have revealed the spatiotemporal variations in the phenological events of woody plants. However, limited studies have focused on the phenophases of herbaceous plants. In this study, by using a meta-analysis method, we extracted information about the phenological changes in herbaceous plants in China's grasslands from existing studies (including the period, station, species, phenophases, phenological trends, and climatic determinants) and analyzed the patterns manifested in the dataset. The results showed that the spring phenophases (e.g., first leaf date and first flowering date) of the herbaceous plants mainly advanced over the past 30 years, but a large difference existed across grassland types. The spring phenophases of forages (species from the Cyperaceae, Gramineae, and Leguminosae families) became earlier in the desert steppe and alpine steppe but showed no apparent trends in the alpine meadow and even became later in the meadow steppe and typical steppe. In most cases, the increase in spring temperatures and precipitation promoted the greening up of herbaceous plants, while sunshine duration was positively correlated with the green-up date of herbaceous plants. For the autumn phenophases, the proportions of the earlier and later trends were very close, but the trends varied among the grassland types. The leaf coloring dates of the forages were delayed in the meadow steppe and alpine steppe but showed no distinct pattern in the typical steppe or alpine meadow and even became earlier in the desert steppe. In most cases, the increase in growing season temperature led to an earlier leaf coloring date of the herbaceous plants, but the increase in the preseason precipitation delayed the leaf coloring date. Our results suggested that the phenophases of herbaceous plants have complicated responses to multiple environmental factors, which makes predicting future phenological changes difficult.