Artificial Intelligence (AI) is revolutionizing various industries, with agriculture being a key area of impact. In agriculture, AI plays a crucial role in improving efficiency, enhancing crop yields, and providing early disease detection, which is essential for minimizing losses. Early identification of plant diseases can significantly enhance food production and ensure food security. In this research, the work utilizes Vision Transformer (ViT) models, which excel in processing image data by capturing global features through Self-Attention Mechanisms (SAM), making them highly effective for disease detection. The model trained and tested on a dataset of Malabar Spinach images demonstrates high accuracy in early disease identification. Essential performance metrics such as F1-score, accuracy, sensitivity, precision, recall, and specificity provide additional validation of the model's effectiveness. This methodology enhances agricultural practices by facilitating timely interventions and promoting improved food production outcomes.