Grapes are a widely cultivated fruit from the genus Vitis, used in fresh consumption, winemaking, and various food products that hold significant economic importance, with the global grape industry valued at over $367 billion in 2023, driven primarily by the wine sector, which alone generated $430 billion in revenue globally. Grape leaf diseases significantly impact the grape industry by reducing yields and quality, costing the global industry an estimated $3 billion annually in lost production. These diseases increase production costs by requiring more frequent fungicide applications, leading to an estimated 20–25% rise in overall cultivation expenses. The integration of AI, deep learning, and image processing is advancing the early and precise detection of grape leaf diseases, facilitating timely interventions that minimize crop losses and enhance the efficiency of pesticide application. This study proposes a parallel-CNN model utilizing MobileNetV2 and Vision Transformer to classify grape leaves which are affected by Black Rot, Leaf Blight, and Black Measles against the healthy leaves. The model was trained, tested, and validated on a dataset of 4,639 images, which were pre-processed using techniques such as image inversion, augmentation, and outlier handling, achieving an accuracy of 99.20% along with a precision, recall, and F1-score of 0.9957. Explainable AI methods, including Grad-CAM and LIME, were utilized to gain deeper insights into the model’s decision-making process. This work has the potential to create a significant social impact by improving disease detection in agriculture, leading to more sustainable farming practices and enhanced food security.