Abstract Objectives: to evaluate the association between dietary intake during pregnancy and different gestational clinical conditions (hypertensive, diabetics, smokers, having intrauterine growth restriction and a control group) and associated factors. Methods: cross-sectional study nested in a cohort study from 2011 to 2016 that occurred in three hospitals in Porto Alegre (Brazil). Sociodemographic conditions and prenatal were investigated and maternal feeding practices were analyzed by the Food Frequency Questionnaire. To calculate the caloric percentage from food groups, food items were categorized into:unprocessed or minimally processed, processed and ultra-processed foods. The Kruskal-Wallis test with Dunn's post-hoc compared food consumption between the groups and the Poisson regression evaluated the association between the variables. Results: there was no statistical difference in food intake among 303 mothers of different gestational clinical conditions, but diabetic pregnant women had lower caloric contribution value of ultra-processed foods. In addition, pregnant women from all groups showed adequate consumption in relation to the percentage of caloric contribution of macronutrients in the total energy value. Conclusions: there was no difference in energy consumption according to different gestational clinical conditions.In diabetic, smokers and hypertensive pregnant women, associations between total energy intake and different sociodemographic factors were observed between the groups.