Adequate diet during pregnancy has positive effects on the mother and pregnancy outcome. Assessment of diet quality during pregnancy is particularly important in areas where household food security is suboptimal, to enable appropriate targeting and intervention. This study assessed diet quality and identified predicting factors among pregnant women in northern Ghana. A cross-sectional study involving 403 pregnant women was conducted in May 2018. Pregnant women attending antenatal care clinics (ANC) were selected using simple random sampling technique. We assessed socio-demographic characteristics, 24-h recall and household food security. The minimum dietary diversity for women (MDD-W) was used as a proxy measure for diet quality based on Food and Agricultural Organization (FAO) guidelines. Logistic regression models were fitted to determine the predictors of diet quality. The mean dietary diversity score (DDS) of 10 food groups was 4.4 ± 1.1 (95% CI: 4.3-4.5). Logistic regression showed that women of high educational level (adjusted odds ratio [AOR] = 2.42; 95% confidence interval [CI] [1.21-4.84]; P = 0.01), women of high household wealth index (AOR = 1.78; 95% CI [1.14-2.77]; P = 0.01], none/mild household hunger (AOR = 2.71; 95% CI [1.26-5.82]; P = 0.01), medium household size (6-15 members) (AOR = 1.66; 95% CI [1.04-2.66]; P = 0.03) and women of gestational age 20-35 weeks (AOR = 1.89; 95% CI [1.05-3.40]; P = 0.03) were more likely to have quality diets after adjusting for potential confounding variables. Diet quality among pregnant women was low and was predicted by educational level, household wealth, gestational age and food security. Women education and improvements in household food security could impact diets of pregnant women in northern Ghana.