The aim of this work was to develop an analytical method to predict total anthocyanins content (TAC) and total phenolic compounds (TPC) in intact wax jambu fruit [Syzygium malaccense (L.) Merryl et Perry] using near-infrared spectroscopy (NIRS) and partial least squares (PLS). The estimation accuracy was based on parameters such as root mean square error of prediction (RMSEP), correlation coefficients [calibration (rc) and prediction (rp) set] and ratio of performance to deviation (RPD). TAC, rp=0.98, RMSEP=9.0mgL−1 and RPD=5.19 were attained using second derivative pre-treatment. TPC, rp=0.94, RMSEP=22.18 (mg gallic acid equivalents (GAE)/100g) and RPD=3.27 (excellent accuracy) were also obtained using second derivative pre-treatment. These findings suggest that the NIRS and PLS algorithms can be used to determine TCA and TPC in intact wax jambu fruit.