Prenatal physical activity (PA) was discussed to decrease the incidence of obstetric and neonatal complications. In this secondary cohort analysis of the cluster-randomized GeliS (“healthy living in pregnancy”) trial, associations between prenatal PA and such outcomes were investigated. PA behavior was assessed twice, before or during the 12th week (baseline, T0) and after the 29th week of gestation (T1), using the self-reported Pregnancy Physical Activity Questionnaire. Obstetric and neonatal data were collected in the routine care setting. Data were available for 87.2% (n = 1994/2286) of participants. Significant differences between the offspring of women who adhered to PA recommendations at T1 and offspring of inactive women were found in birth weight (p = 0.030) but not in other anthropometric parameters. Sedentary behavior was inversely associated with birth weight at T1 (p = 0.026) and, at both time points, with an increase in the odds of low birth weight (T0: p = 0.004, T1: p = 0.005). Light-intensity PA at T0 marginally increased the odds of caesarean section (p = 0.032), but neither moderate-intensity nor vigorous-intensity activity modified the risk for caesarean delivery at any time point. The present analyses demonstrated associations between prenatal PA and some neonatal and obstetric outcomes.