Chronic Kidney Disease (CKD) is a global health issue with high morbidity and death rates. Early and accurate diagnosis is critical to slowing disease progression and improving patient outcomes. Disease diagnosis is the analysis of the biochemical and physiological causes of disease. It is one of the most complex and precise processes physicians perform daily since it is the basis for many critical medical decisions regarding patients’ health. A massive volume of clinical data is being produced daily, and the information provided by each kind of this data must be evaluated and assigned for the diagnostic processes. A sophisticated system is needed to help physicians diagnose the disease accurately. In this study, we utilized an effective and efficient chronic kidney disease diagnosis prediction utilizing various machine-learning models. The Kaggle dataset is used in this study. Synthetic Minority Oversampling Technique (SMOTE) is implemented to balance the dataset. The models implemented are K-nearest neighbor (KNN), logistic regression, random forest, support vector machine (SVM), gradient boosting, XGBoost, AdaBoost, and CatBoost. The performance reveals that we have acceptable results for all models, but SVM is the best because of the greatest accuracy (99.16%). At the same time, the CatBoost Classifier and the XGB Classifier are deemed promising with accuracy equal to 97.7% and 97.28%, respectively.