In this study, our objective was to determine whether plant stoichiometry following nitrogen (N) and phosphorous (P) enrichments can enhance understanding regarding biodiversity loss processes. Thus, we conducted a field experiment involving N, P, and N + P enrichments in a sub-alpine meadow on the northeast of Qinghai-Tibet Plateau within the 2009–2014 period. Plant stoichiometric patterns were investigated using six exemplar species, four functional group levels, as well as the community level. The hierarchical responses of species richness, dominance, plant height, and plant ecological stoichiometry to nutrient addition were investigated. Further, step-wise regression analysis was performed and a structural equation model (SEM) was constructed to assess the linkages between ecological performances and plant stoichiometric characteristics. The model was also used to quantify the contributions of stoichiometric characteristics as drivers of changes in species richness. Our results indicated that N-only and N + P enrichments significantly reduced species richness at community-level, while enhancing grass richness and dominance. Furthermore, only the responses of the N:P ratio, P-related stoichiometry (P concentration, C:P ratios, and N:P ratio), and plant C:N:P stoichiometry to N, P, and N + P fertilizations, at community level could be mirrored at functional group and species levels. Additionally, the responses of the plant N contents and C:N ratios of different species and functional groups showed 28% similarity following nutrient addition. However, at functional group level, only 10% of these responses were similar to those observed at community level, and only in grasses and at community level did the stoichiometric responses show any convergence. We also observed that the major factors affecting ecological performance at different levels included plant N contents and C:N ratios. The species dominance of Elymus nutans (E.n) and the functional group dominance of grasses increased with increasing plant N concentration, and in terms of species richness, species, functional group, and community level stoichiometric traits could fully explain 43% of the observed species richness trend. Further, the effect of the community C:N ratio on species richness was significantly positive. Those results indicated stoichiometric responses to N and P fertilization did not entirely show scale-independence. Moreover, in a N-limited community, plant differences with respect to nitrogen content flexibility and nitrogen use efficiency can serve as key drivers in explaining ecological performance. Community C:N ratios showed suitability for used in the prediction of species richness trends, and to clarify changes in biodiversity, it is necessary to investigate stoichiometric traits at functional group level.