Breast cancer is the most commonly diagnosed cancer and the fifth leading cause of death in women. Early detection of this disease not only increases the survival rate but also reduces the cost of treatment. Mammography (X-ray mammography) is the current imaging method to identify and diagnose breast malignancies. In this work, we propose a classification technique based on several network architectures, including NasNetLarge, MobileNetV2, InceptionV3, DenseNet, and Vision Transformer to classify mammograms as normal, benign, or malignant. Experimental results show that the accuracy of the proposed models is up to 99%. The support of mammograms screening containing lesions will help doctors focus more on analyzing results. This helps the accuracy of diagnosis to increase and gives timely treatment direction.