Highbush blueberries (Vaccinium corymbosumL.) are cultivated worldwide for their fruit with unique tasteand potential health benefits. Blueray, Bluecrop, and Spartanare prominent among the various blueberry cultivars. Weperformed gas chromatography–mass spectrometry (GC–MS)-based metabolic profiling to differentiate the fruits ofthese three cultivars, and built an optimal partial leastsquares-discriminant analysis (PLS-DA) model to separatethem. Amino acids, fatty acids, organic acids, phenoliccompounds, and sugars were identified in the fruits. Theoptimized PLS-DA model for different cultivars of the fruitswas obtained by selecting variables based on a variable importancein the projection (VIP) cut-off value of 1.0. Caffeicacid, aspartic acid, acetic acid, threonolactone, inositol, xylose,glucoside, linolenic acid, mannose, altrose, glycinealanine, and valine were found to be relevant and contributingcompounds for differentiating cultivars. In addition, ahierarchical cluster analyses dendrogram pattern was correlatedwith the PLS-DA. This study suggested that GC–MSbasedmetabolic profiling coupled with multivariatestatistical analysis could be used to differentiate the fruits ofthree major highbush blueberry cultivars.