Abstract Background Dietary intake patterns of children from the 2004 Pelotas birth cohort study have been described at 12, 24 and 48 months of age, but there is no information about dietary patterns of these children at 6 years. Then, we aimed to identify and describe dietary intake patterns of children aged 6 years as well as to assess their association with socioeconomic and demographic characteristics, early feeding practices and BMI z-score at 6 years. Methods We used principal components analysis to identify dietary intake patterns of 3,427 children from the 2004 Pelotas (Brazil) birth cohort study. We used multiple linear regression models to evaluate whether socioeconomic and demographic characteristics (socioeconomic position, mother’s age at birth, and child’s sex and skin colour), early feeding practices (exclusive breastfeeding duration and age of introduction of complementary foods), and BMI z-score at 6 years were associated with dietary intake patterns. Results We identified seven dietary components of children’s dietary intake patterns, namely: fruits and vegetables, snacks and treats, coffee and bread, milk, cheese and processed meats, rice and beans and carbohydrates. Dietary patterns were socially patterned, since six dietary components were associated with socioeconomic position. Moreover, high intake of snacks and treats and less fruits and vegetables were associated with children born to teenage mothers, with those exclusively breastfed for less than one month, and with those who started on complementary feeding before 4 months. Finally, overweight and obese children at 6 years presented lower intake of four out of seven dietary components, but we need to be cautious in interpretation due to limitations on food consumption reporting and due to possible reverse causality. Conclusion Dietary intake patterns in children are strongly influenced by socioeconomic characteristics. Other factors such as younger maternal age at birth, and both early weaning and early introduction of complementary feeding appear to be related with ‘unhealthier’ patterns. Overweight and obese children presented lower intake of four out of seven dietary components, but further studies would be interesting to understand the longitudinal effect of children’s feeding practices on BMI and adiposity.