India is a land of herbal plants, which have been utilized for various purposes because of their active components in treating many diseases. These plants’ leaves, stems, flowers, and fruits are used in distinct ways. To get the full benefits of these plants, deep learning (DL) technology has been widely employed to automatically identify them using leaf images, which are always available. It employs convolutional neural networks (CNNs), mostly used for image classification. The study focused on classifying ten different classes of herbal plant leaves using three pre-trained CNN transfer learning models—ResNet50V2, MobileNetV2, and DenseNet201. These models have been evaluated using various metrics such as accuracy, precision, recall, and F1-score, demonstrating the ensemble model’s usefulness.