This study modeled the potential distribution of endemic Potaninia mongolica in the eastern Central Asian desert by using ecological niche model: the Genetic Algorithm for Rule-set Prediction (GARP). We created predictive models using herbarium collection data and environmental layers. For model evaluation, we chose the area under the receiver operating characteristic (ROC) curve (AUC) as a single measure. Our results showed that: (1) GARP obtained high scores of performance (AUCs >; 0.9); (2) the visual inspection of model results indicated that the generated most potential suitable areas for P. mongolica covered the most desert areas, specifically including the significant portions of southern Mongolia, such as Alashan Gobi, and the East Gobi; the Alashan Left and Right Banner as well as the northern areas of Inner Mongolia Autonomous Region; and the western Yumen, the middle Hexi Corridor of the Gansu Province; (3) the ecological niche of P. mongolica was defined in a widely arid environmental space: Max Temperature of Warmest Month were almost during 30.2-36.8 °C, Min Temperature of Coldest Month were ranged from -15.6 °C to -16.9 °C, and the Annual Precipitation were between 40-80 mm.