In the present study, carotenoids, anthocyanins, and phenolic acids of cauliflowers ( Brassica oleracea L. ssp. botrytis) with various colored florets (white, yellow, green, and purple) were characterized to determine their phytochemical diversity. Additionally, 48 metabolites comprising amino acids, organic acids, sugars, and sugar alcohols were identified using gas chromatography-time-of-flight mass spectrometry (GC-TOFMS). Carotenoid content was considerably higher in green cauliflower; anthocyanins were detected only in purple cauliflower. Phenolic acids were higher in both green and purple cauliflower. Results of partial least-squares discriminant, Pearson correlation, and hierarchical clustering analyses showed that green cauliflower is distinct on the basis of the high levels of amino acids and clusters derived from common or closely related biochemical pathways. These results suggest that GC-TOFMS-based metabolite profiling, combined with chemometrics, is a useful tool for determining phenotypic variation and identifying metabolic networks connecting primary and secondary metabolism.