To develop a quantitative models for simultaneous determination of pinoresinol diglucoside (PDG) and geniposidic acid (GPA) in Eucommia ulmoides with near-infrared (NIR) spectroscopy combined with chemometrics. The NIR spectra were collected in diffuse reflection mode and pretreated with various spectra preprocessing methods including first-order differentiator, multiplicative scatter correction and so on. The optimal wavelength variables were screened out by competition adaptive weighted sampling method. The quantitative models for the simultaneous determination of PDG and GPA in Eucommia ulmoides were established with partial least squares (PLS) algorithm and cross validation methods. The quantitative prediction models for simultaneous determination of PDG and GPA in Eucommia ulmoides showed good predictive ability. The correlation coefficients (R2) of the two calibration models were 0.961 5, 0.958 3 while the roots mean square of cross-validation (RMSECV) were 0.001 5, 0.006 4, respectively. The quantitative prediction models proved that near infrared spectra method used for the quantitative analysis of PDG and GPA in Eucommia ulmoides owned high prediction accuracy and can meet the precision need of rapid determinations of PDG and GPA in Eucommia ulmoides in reality so t it provides a new method to realize the real time on line of quality control of Eucommia ulmoides.