Dynamic assessment is currently discussed in educational literature as one of the most promising practices in stimulating learning among various groups of students, including gifted and potentially gifted students. The present study investigates effects of dynamic assessment on mathematics achievement among elementary school students, with potential giftedness analyzed as additional predictor. Two samples of primary school students participated in a quasi-experimental study: the experimental condition consisted in application of dynamic assessment procedures, contrasted with regular classroom assessment in the control condition. Math achievement was measured with curriculum-based tests, and potential giftedness was estimated based on nomination scores assigned by classroom teachers and parents for each child. Results suggest that dynamic assessment procedures produce significant effects on math achievement among elementary school students, and potential giftedness enhances this effect. Study limitations and educational implications are discussed, and further research paths are indicated.