Phytoplankton play a crucial role in biogeochemical cycling and aquatic food webs while also susceptible to environmental variations. However, their response to altitude gradients remains poorly understood. In this study, we applied a metabarcoding approach to explore eukaryotic phytoplankton community structure, co-occurrence networks, and assembly processes along a steep altitudinal gradient (590-4,500 m) in the Nyang River and the lower reaches of the Yarlung Zangbo River on the Qinghai-Tibetan Plateau during dry and wet seasons. Using 18S rDNA sequencing, we obtained 2,852 amplicon sequence variants. Our results demonstrated that Ochrophyta was the dominant taxon in the eukaryotic phytoplankton community across both seasons. Alpha diversity exhibited distinct seasonal patterns, decreasing monotonically with increasing altitude in the dry season whereas the highest diversity was observed at medium altitudes in the wet season. Phytoplankton co-occurrence networks became more topologically complex as species diversity increased. Among environmental factors, altitude (r = 0.62), water temperature (r = 0.52) and pH (r = 0.51) significantly influenced phytoplankton communities. Stochastic processes globally dominated phytoplankton community assembly (66%) and became increasingly influential from dry season (51%) to wet season (71%). Their impact gradually increased from low altitude (57%) to medium altitude (64%), but deterministic processes overwhelming dominated community assembly at the higher altitude in both seasons (dry season: 95%, wet season 71%). In summary, these findings enhance our understanding of the spatial and temporal dynamics of eukaryotic phytoplankton communities in highland rivers and the maintenance of planktonic diversity along elevational gradients.