Abstract: Global warming exacerbates the intensity, frequency, and duration of extreme climate events, and has profound impacts on vegetation phenology. Plateau mountains are highly sensitive to climate change, resulting in a more complex response mechanism. In this study, we utilized multi-source remote sensing data to identify key vegetation phenology periods, namely the Start of Season (SOS), the End of Season (EOS), and the Length of Season (LOS), and derived 17 extreme climate indices from national meteorological station data. A comprehensive suite of analytical methods was utilized, including linear regression, Sen’s slope, Mann-Kendall significance tests, partial correlation analysis, and structural equation modeling, to quantitatively assess the spatiotemporal variations in vegetation phenology and its response mechanism to extreme climate events with varying intensities, frequencies, and durations in Southwest (SW) China. The results show that: 1) during 2003–2022, SOS has advanced by 0.22 d/yr, EOS has delayed by 0.28 d/yr, and LOS has lengthened by 0.49 d/yr. Phenology shifts are altitude-sensitive, with a later SOS, an advanced EOS, and a shortened LOS at higher elevations. SOS begins earliest in the central and north subtropical zones of study area, including eastern Sichuan, Chongqing, Guizhou, and Yunnan, and latest in the plateau frigid zone, including northern Xizang. EOS ends earliest in the plateau frigid zone and latest in the marginal tropics and south subtropical zones, with the plateau frigid zone also having the shortest LOS. 2) Temporally, extreme precipitation intensity and frequency in the region have risen significantly, with dry-related event duration shortening and wet-related duration lengthening. Extreme warming prevails, with cold extremes decreasing and high-temperature extremes increasing significantly. The intensity of extreme precipitation extends to high-altitude areas, with the overall spatial coverage expanding, and the most significant changes observed in Xizang. The frequency of extreme precipitation gradually shifted and concentrated eastward, showing more prominent increasing characteristics in Sichuan and Chongqing. Meanwhile, consecutive wet days (CWD) exhibits distinct spatial distribution characteristics, mainly concentrated in eastern Sichuan and parts of Chongqing and Guizhou. 3) SOS negatively correlates with extreme precipitation intensity and duration indices but positively with frequency indices, indicating enhanced extreme precipitation promotes vegetation growth and advances SOS. SOS is positively correlated with warm spell duration (WSDI), while EOS shows the strongest negative correlation with number of warm days (TX90p). These results suggest extreme temperature indices inhibit vegetation growth initiation and cessation, with extreme precipitation as the primary driver of regional vegetation phenological changes. These findings deepen the understanding of vegetation phenology’s response to extreme climate events in SW China’s plateau mountains and provide scientific support for optimizing ecological conservation and climate adaptation strategies in sensitive areas.