The study investigated the predictive value of robust and specific personality traits in adolescents (M[subscript age]?=?14.7 years), in explaining their academic achievement at the end of basic compulsory schooling. Personality data were obtained through self, maternal, and peer reports using the Inventory of Child/Adolescent Individual Differences. Adolescent gender and maternal education predicted 36, 26, 19, and 26% of the variance in the final grades in Slovene, English, and mathematics and the overall GPA, respectively. Personality ratings by each of the three groups of informants substantially improved the prediction of students' academic achievements, over and beyond gender and maternal education. The robust trait scores contributed to significant increments in the variance explained, across the academic achievement indicators, ranging from 8 to 17% (self-report), 15 to 24% (maternal report), and 20 to 32% (peer report). Conscientiousness was consistently the most powerful predictor of students' academic success and extraversion was negatively associated with all achievement indicators. The study provided support for a relatively stronger predictive utility of specific, rather than robust personality traits. Likewise, peer ratings of the students' personality provided relatively larger increments in variance explained in academic achievements than maternal and self-ratings; in general, the personality trait ratings added more to the prediction of mathematics grades and the GPA relative to the prediction of success in languages. Among specific traits, subjectively perceived student intelligence was the most powerful and consistent predictor of final grades and GPA.