An earthquake can prove to be an extremely destructive disaster that endangers life and property in many parts of the world, hence predicting earthquakes adequately earlier can help us a great deal in keeping ourselves and our properties protected from damage. Earthquake magnitude prediction is a complex task because of the unpredictable nature of seismic activities. The ability to accurately predict the earthquake’s magnitude is crucial for disaster management and risk assessment. Conventional prediction techniques often lack the precision and reliability required for effective mitigation strategies. We address this challenge by using different regression and boosting models such as decision tree, extra tree, random forest, ADABoost, and XGBoost for the magnitude prediction. We will use these models on our dataset and measure each model’s performance by calculating its mean squared error, mean absolute error, and R-squared value. By comparing the performances of these models, we will find the most suitable model for earthquake magnitude prediction. We expect our models to yield accurate predictions of earthquake magnitudes, and, moreover, to perform well in a variety of datasets and seismic locations, demonstrating their dependability in real-world applications. Our analysis revealed that out of the regression and boosting models used, random forest regressor achieved the highest accuracy of 99.68%.