Ecoenzymatic stoichiometry is an essential method for predicting material cycling, nutrient limitations, and balance within ecosystems, provides new insights into microbial metabolic mechanisms. However, the status and drivers of soil microbial nutrient limitation in alpine grasslands remain unclear. In this paper, soil samples were collected along a seven-altitude gradient (2400 m to 3000 m). The vector analysis model, vector-threshold (V-T) model, and threshold model were used to predict microbial nutrient limitations in the Bayinbuluke grassland. The driving factors influencing soil microbial nutrient metabolism in the alpine grassland ecosystem were examined including soil, plants, and microorganisms. The results from the three models indicated that soil microorganisms in this region primarily experienced microbial phosphorus limitation. The threshold model demonstrated greater accuracy in predicting microbial nutrient limitation, compensating for the shortcomings of the vector and threshold element ratio models. Through partial least squares model, it was found that soil nutrients and pH were significant factors affecting soil microbial phosphorus limitation. Through the results of this study, it can be seen that it is scientific and accurate to predict microbial nutrient limitation by model. The findings of this study are important for enhancing the understanding of nutrient balance and its driving mechanisms within the alpine grassland ecosystem in Xinjiang.