The goal of this study was to describe the relationship between the creative abilities and the school grades of high school students in Poland. Almost six hundred (N = 589) students from 34 high schools from all over Poland participated in the study. Their creative abilities were measured by using the Test of Creative Thinking-Drawing Production (TCT-DP), and the school results were measured by GPA. Students' intelligence level (as measured by the Raven's Progressive Matrices) and their gender were controlled. The analyses were based on OLS regressions as well as on multilevel models controlling for grouping students into classes. It was shown that creative abilities are not correlated with students' GPA, yet the multilevel control of grouping students into classes demonstrated interesting and potentially important differences. In some schools, the relations were positive, strong and statistically significant, while in others they were non-existent or negative. The role of creative abilities for GPA was greater in larger schools and in schools located in big cities. We discuss the possible reasons for and consequences of our findings. (Contains 3 tables and 5 figures.)