Introduction: Low adiponectin level is an indication of insulin resistance during gestational weight gain, and could be used as a marker for diabetes or hypertension in pregnancy. Objective: To find the relationship between BMI and adiponectin levels in pregnant women. Methods: In this cross-sectional study, total 200 study participants were selected. Participants were 24 to 40 weeks pregnant, with or without gestational diabetes mellitus. Patients with pre-existing or pre-pregnancy diabetes were excluded from study. BMI was calculated at the time of the study. Adiponectin levels were measured on Metrolab ELISA, using Human Adiponectin ELISA kit by Biovendor, Germany. SPSS version 20 was used to analyze data. Results: Out of 200 participants, 100 had gestational diabetes. None of the participants was below normal weight, however, about 38% of participants were overweight, and about 58% of participants were obese, according to WHO-Asian criteria. Adiponectin level was below normal in more than 60% of patients. No difference was found between BMI within diabetes-based groups. But there was a significant difference between adiponectin levels in Group I and Group II. No relationship of BMI was found with diabetes-based groups, gestational age groups, or adiponectin categories. A significant relationship was found between the diabetic group and low serum adiponectin levels. Regression analysis showed that BMI was neither a predictor of diabetes nor adiponectin levels. However, adiponectin levels were a predictor of diabetes in pregnancy. Conclusion: Our study could not find a relationship of BMI with either adiponectin levels or gestational diabetes. Low adiponectin levels, however, could predict gestational diabetes.