Education serves as the foundation for cultivating a capable and advanced society. Enhancing student performance directly contributes to strengthening the education system and, in turn, shaping a better future for society. This highlights the critical importance of forecasting student outcomes, an area that continues to garner significant interest. Leveraging advancements in machine learning, our research aims to build an efficient model for predicting student grades. We utilized datasets collected in 2019 from students of the Faculty of Engineering and the Faculty of Education at the University of California, Irvine. These datasets encompass eight distinct grade categories. A group of machine learning algorithms called ensemble learning techniques was used to build our model. These included gradient-boosted trees, histogram-based gradient boosting, extreme gradient boosting, light gradient boosting machines, categorical boosting, and extremely randomized trees. Furthermore, we utilized other ensemble learning techniques, like voting and stacking, to develop new models based on the previous ones. By evaluating metrics, such as accuracy, F1 macro-score, and training/testing time, we identified the most efficient and fastest performing models.