Breast cancer detection is a critical area of medical research where early and accurate diagnosis can significantly improve patient outcomes. This study conducts a comprehensive comparative analysis of various deep-learning models for detecting Invasive Ductal Carcinoma (IDC), the most prevalent subtype of breast cancer. We evaluate the performance of Convolutional Neural Networks (CNNs), MobileNet, Transformer, and EfficientNet models using a comprehensive dataset of IDC images. The models are assessed based on key metrics, including accuracy, F1-score, precision, recall, and training time. Our results indicate significant differences in performance metrics, with EfficientNet achieving the highest accuracy and demonstrating exceptional reliability and precision. MobileNet also performs strongly, providing a balance of high accuracy and efficiency. These findings offer valuable insights into the suitability of different deep learning architectures for breast cancer detection, guiding future research and clinical applications toward more effective diagnostic tools.