Objective To develop and validate a risk prediction model for Gestational Diabetes Mellitus （GDM）based on demographic characteristics，clinical features and biochemical indicators. Methods A self ⁃ administered questionnaire and the hospital electronic medical record system were used to collect data. A prediction model for GDM was established using logistic stepwise regression analysis. The area under curve（AUC），the calibration curve and decision curve analysis were used to evaluate the discrimination，the consistency and clinical usability of the model respectively. A nomogram was performed.Results The AUC of a GDM prediction model based on age，history of GDM，mean arterial pressure and HbA1c constructed through logistic regression analysis was 0.75（95%CI：0.71 ~ 0.80）. The calibration curve showed good consistency. By decision curve analysis，the model showed a positive net benefit when the threshold probability ranged between 0.05 ~ 0.75. Conclusion The GDM prediction model has good predictive performance，and provides an effective assessment tool for GDM.