This study proposes the use of Convolutional Neural Networks (CNN) for classifying brain tumor types, specifically Glioma, Meningioma, Pituitary, and healthy brain MRI images. The dataset used consists of 7023 MRI images, including 1621 Glioma, 1645 Meningioma, 1757 Pituitary, and 2000 healthy brain images. Two CNN architectures were developed, referred to as Scheme 1 and Scheme 2. The results showed that Scheme 2 achieved an F1-Score of 98% and an Accuracy of 99%, outperforming Scheme 1 (F1-Score 96%, Accuracy 98%). These results are higher than several previous studies, which reported accuracy values of 94.39%, 97.54%, 97.18%, 96.08%, 96.36%, and 95.55%. The study also developed a simple API to assist medical professionals in diagnosing brain tumors from MRI images, achieving a diagnosis in under 3 seconds. The results indicate that the proposed CNN model is effective in classifying brain tumors with high accuracy and efficiency.This study proposes the use of Convolutional Neural Networks (CNN) for classifying brain tumor types, specifically Glioma, Meningioma, Pituitary, and healthy brain MRI images. The dataset used consists of 7023 MRI images, including 1621 Glioma, 1645 Meningioma, 1757 Pituitary, and 2000 healthy brain images. Two CNN architectures were developed, referred to as Scheme 1 and Scheme 2. The results showed that Scheme 2 achieved an F1-Score of 98% and an Accuracy of 99%, outperforming Scheme 1 (F1-Score 96%, Accuracy 98%). These results are higher than several previous studies, which reported accuracy values of 94.39%, 97.54%, 97.18%, 96.08%, 96.36%, and 95.55%. The study also developed a simple API to assist medical professionals in diagnosing brain tumors from MRI images, achieving a diagnosis in under 3 seconds. The results indicate that the proposed CNN model is effective in classifying brain tumors with high accuracy and efficiency.