Objective To construct a dynamic interpretable prediction model for oral frailty risk in community-dwelling older adults, and to provide a reference for community nurses to carry out early assessment and intervention. Methods Using convenience sampling, 1,457 community-dwelling older people in Anhui Province were recruited and assessed using the Oral Frailty Index-8 by trained investigators.The relevant influencing factors for oral frailty were collected as potential variables for the model.Participants were randomly divided into a training set (n=1,020) and a validation set (n=437) at a 7∶3 ratio.Based on the training set, inf-luencing factors were identified through 8 machine learning algorithms, including logistic regression, random forest, support vector machine, decision tree, neural network, extreme gradient boosting, naive Bayes, and K-nearest neighbors.Model performance was evaluated using the validation set.The optimal model was interpreted and visualized using SHapley Additive exPlanations (SHAP). Results The incidence of oral frailty risk was 64.4% in the community-dwelling older adults.The support vector machine model had the best predictive performance in the validation set, with an AUC of 0.783, F1 score of 0.659 (the highest）, and Brier score of 0.180 （the lowest）; chronic disease, age, depression, smoking, malnutrition, and physical frailty were the main influencing factors of oral frailty.SHAP results showed that chronic disease of the greatest contribution. Conclusion The support vector machine model has satisfactory predictive performance for the risk of oral frailty in community-dwelling older adults.The main risk factors identified based on SHAP can provide a basis for the development of targeted nursing interventions.