 This paper proposes a novel approach for classifying brain tumors based on MRI images. It combines the strengths of three different models, a convolutional neural network, CNN, without augmentation, a CNN with augmentation, and a transfer learning VGG-19 model into one ensemble model. This model is then combined with a weighted average algorithm to select the best combination of weights, which are determined through a grid search. The resulting model outperformed all three individual models in terms of accuracy, precision, and F1 score. Thus, the proposed model can serve as a useful tool for radiologists to quickly and accurately diagnose brain tumors from MRI images. This article was authored by Vatsala Anand, Shifli Gupta, Dipali Gupta, and others.