Coffee ranks among the most popular beverages and is a widely traded commodity worldwide, with approximately 30% to 40% of the global population consuming coffee daily. Given the rising demand and the challenges posed by climate change and nutrient imbalances, early detection of coffee leaf diseases has become increasingly essential for farmers and the industry. One prevalent issue is Coffee Leaf Rust, which significantly impacts coffee plants and can lead to substantial crop losses. Although there have been considerable advancements in detection and classification techniques, traditional methods relying on naked-eye observation remain common. This subjective approach often lacks the reliability and precision needed for effective disease management. To achieve more accurate and consistent results, it is necessary to adopt objective methods that enable timely detection and precise classification, ultimately leading to effective treatment of the disease. This study proposes using the YOLOv8n-cls model, a fast single-stage object detection and classification tool, to classify Coffee Leaf Rust in the RoCoLe dataset, which consists of images of Robusta coffee plants. The goal is to enhance disease detection accuracy, support regular interventions, and ultimately safeguard coffee plant health and productivity.