Cross-linguistic research has shown that boundaries for lexical categories differ from language to language. The aim of this study is to explore these differences between languages in relation to the categorization differences within a language. Monolingual Dutch- (N=400) and French-speaking (N=300) Belgian adults provided lexical category judgments for three lexical categories that are roughly equivalent in Dutch and French. Each category was represented by good, borderline, and bad examples. A mixture modeling approach enabled us to identify latent groups of categorizers within a language and to evaluate cross-linguistic variation in relation to within-language variation. We found complex patterns of lexical variation within as well as between language groups. Even within a seemingly homogeneous group of speakers sharing the same mother tongue, latent groups of categorizers display a variability that resembles patterns of lexical variation found at a cross-linguistic level of comparison.