Coffee is a significant commercial crop that is grown throughout the world, and it is second only to crude oil in terms of trade volume. For many farmers, it serves as their primary source of daily income. Thus, the primary issues influencing agricultural and economic output in many nations are controlling coffee leaf diseases and ensuring the quality of coffee bean products. One of the most well-known diseases affecting coffee leaves is Rust, followed by Phoma, Cercospora, and Miner. For disease detection and identification, farmers and professionals often use their unaided eyes to observe the plants. However, this strategy could be time-consuming, expensive, and unreliable. Due to the rising interest in using deep learning in farming, numerous studies have demonstrated that image classification is very reliable in recognizing plant diseases. Over the past few years, researchers have attempted to produce deep-learning solutions for cultiva- tion in terms of disease and species classification using convolutional neural networks (CNN). Therefore, we suggested a framework called the Coffee Leaf Quadruple Classifier (CLQC) that is divided into three unique stages each stage contains four models. They are VGG16, Efficient- NetB0, DenseNet121, and RestNet152V2 selected due to their accurate classification in coffee leaf diseases. The evaluated results indicate that by using deep convolutional models on a set of data, pre-processing, and a variety of supervised deep learning strategies across three distinct phases, the EfficientNetB0 model outperformed other models in all three stages. It achieved 99.91% accuracy in the first stage, 99.45% accuracy in the second stage, and 99.95% accuracy in the third stage.