Despite the systematic recognition of large gender gaps in favour of boys in mathematics performance in Chile, the role of schools in explaining this gap has been comparatively underresearched. This paper analyses a longitudinal dataset to explore within- and between-school variation of Chilean primary schools in promoting attainment and progress of girls and boys. In order to reflect the hierarchical nature of the educational data, multilevel modelling is used, with 163,044 students, nested within 3,355 schools, within 310 municipalities to fit raw, contextualised attainment, and value-added models. The weak evidence indicative of gender differential school effect in progress in Chilean primary schools was interpreted as non-substantive. However, differences in effectiveness played a significant role, as girls progressed more than boys in less effective schools, but the opposite was true in more effective schools. Finally, the study concludes that the Chilean gender gap needs to be addressed mainly beyond schools.