Research suggests that lowering maternal morbidities associated with gestational diabetes mellitus (GDM) can be achieved with earlier risk group identification. Therefore, the purpose of this study was to examine potential markers for identifying first-trimester pregnant women who are at high risk for developing GDM. This was a retrospective cohort study. The pertinent maternal clinical data were retrieved prior to 13+6 weeks of gestation, and a binary logistic regression analysis was used to identify potential GDM predictors. The predictive accuracy was evaluated using the area below the receiver operating characteristics curves. In comparison to the control group, the GDM group had significantly higher mean values for age, body mass index (BMI), mean fasting blood glucose (FBG), and hemoglobin (p < 0.05). The Pearson's correlation coefficients indicated that the first-trimester FBG was significantly positively correlated with the second-trimester FBG. Higher FBG and BMI values were associated with an increased risk of developing GDM (odds ratio [OR] = 3.04, 95% confidence interval [CI] = 2.03-4.55 and OR = 1.18, 95% CI = 1.12-1.25). In terms of predicting GDM, the FBG parameter demonstrated the greatest area under the curve values (0.66), followed by the BMI parameter (0.69). For GDM prediction, the cut-off value for FBG was 4.32 mM, whereas that for BMI was 23.7 kg/m2. The first-trimester FBG and BMI could be utilized to predict gestational diabetes.