Wheat is the third most harvested and consumed grain in the world. A large amount of the wheat harvest spoils due to diseases. Wheat crops are susceptible to more than two dozen diseases. As a result, it becomes extremely difficult to diagnose certain diseases manually. Early detection of wheat leaf diseases is important for ensuring optimal crop yield. Traditional diagnostic methods are frequently time-consuming, requiring expert knowledge, which highlights the need for automated solutions. This study presents an explainable AI (XAI) framework for wheat leaf disease prediction using machine learning models within the ORANGE data mining tool. Neural Network, Support Vector Machine, and Logistic Regression Models are compared for prediction. Wheat leaf image dataset is used in this study. The dataset is categorized in three categories. There are 102 images that are in the healthy category, 208 images that have stripe rust category, and 97 images that have septoria category. ORANGE was used to preprocess images, extract features via embedding, and apply dimensionality reduction techniques like PCA and t-SNE. The study identifies the most effective models based on accuracy and explainability metrics, demonstrating how XAI can enhance transparency in agricultural machine learning predictions.