Wheat leaf disease is a major concern in agriculture. It leads to significant crop yield losses. It is necessary to diagnose wheat leaf disease in its early stages to ensure food security and sustain global wheat production. The main objective of this paper is to present a different method for wheat disease detection using you only look once algorithm version 8 (YOLOv8) and generative adversarial networks (GANs). YOLOv8 is a famous object detection method which can detect and classify objects in real time accurately. It can process the images very quickly and accurately, thus making it an ideal choice for this task. One major problem is limited training data for various wheat diseases. To address this problem, in the proposed research, the authors have introduced a conditional-generative adversarial network (C-GAN)-based data augmentation technique which generates synthetic images of wheat leaves. This technique increases the volume of dataset to be used in training, thus improving the overall generalization of the model. The proposed model further compares the YOLOv8-trained model with other existing models. The proposed model achieves an accuracy of 99.8%, which is better than other models.