Oryza sativa, often known as rice, is a staple food for around half of the world’s population, especially in Asia and Africa. As a staple crop in South Asia, timely and precise classification of rice leaf diseases is essential to maintaining food security. This paper proposes an efficient and explainable ensemble framework for automated rice leaf disease classification using lightweight deep learning models. A comprehensive image dataset of seven rice leaf diseases was created by combining and preprocessing samples from various sources, then balancing the classes to correct dataset imbalance. We benchmarked compact architectures (MobileNetV3Large, EfficientNet-B0, RegNet400 MF, MobileViT-S) on a curated dataset, with RegNetY-400MF achieving 93.98% accuracy and a soft-voting ensemble reaching 94.18%. We used Grad-CAM techniques on the models to make them easier to understand. These visualizations showed which areas were most important for making disease predictions. The proposed framework is highly efficient, generalizes well across data splits, and is readily adaptable for real-world agricultural deployment. This work provides a reproducible baseline for lightweight, scalable, and explainable crop disease diagnostics for sustainable agriculture.