In the domain of image processing, deblurring remains a challenging task, particularly in restoring high-fidelity details from blurred images. This paper introduces a novel approach that synergizes the strengths of Generative Adversarial Networks (GANs) and Transformer models to address this challenge. Our methodology leverages the generative capability of GANs to produce realistic and sharp images, while incorporating the Transformer’s prowess in capturing long-range dependencies and intricate details within images. This adversarial training regime is designed to refine the model’s ability to generate deblurred images that are not only accurate in detail but also high in perceptual quality. Through extensive experiments on various datasets, our model demonstrates superior performance in deblurring, outperforming existing state-of-the-art methods in terms of both objective metrics and subjective visual quality.