Bone fractures are prevalent injuries that necessitate prompt and precise diagnosis to ensure effective treatment. Traditionally, fractures are diagnosed using data from X-rays, magnetic resonance imaging (MRI), or computed tomography (CT) scans. This study aims to assist physicians by employing artificial intelligence to classify X-ray images as fractured or non-fractured. To achieve this, we evaluated the performance of 20 models. The models tested include AlexNet, ResNet, DenseNet, VGG, MobileNet, EfficientNet, RegNet, and RepVGG. DenseNet, MobileNet, and EfficientNet performed best, achieving a test F1-Score of 0.79.