Treatment duration is one of the most important factors that patients consider when deciding 20 whether to have orthodontic treatment or not. This study aimed to build and compare Machine 21 Learning (ML) models for prediction of orthodontic treatment length and to identify factors affect-22 ing the duration of orthodontic treatment using the ML approach. Records of 518 patients who suc-23 cessfully finished orthodontic treatment were used in this study. 70% of the patient data was used 24 for training ML models, and 30% of data was used for testing these models. We applied and com-25 pared nine machine-learning algorithms: Simple Linear Regression, Modified Simple Linear Regres-26 sion, Polynomial Linear Regression, K Nearest Neighbor, Simple Decision Tree, Bagging Regressor, 27 Random Forest, Gradient Boosting Regression, and AdaBoost Regression. We then calculated the 28 importance of patient data features for the ML models with the highest performance. The best over-29 all performance was obtained through Bagging Regressor and AdaBoost Regression ML methods. 30 The most important features in predicting treatment length were age, crowding, artificial intelli-31 gence case difficulty score, overjet, and overbite. Without patient information, several ML algo-32 rithms showed comparable performance for predicting treatment length. Bagging and AdaBoost 33 showed the best performance when patient information, including age, malocclusion, and crowd-34 ing, was provided.