The deep gaze of humans into the night sky, aimed at uncovering the mysteries of the cosmos using advanced telescopes, has generated an immense volume of data. The classification of stars and galaxies present in these images, due to the vast amount of data, was a highly time-consuming process for astronomers. As a result, the "Galaxy Zoo" citizen science project, in which volunteers participated in the classification of this data, was introduced by researchers, significantly reducing the classification time. In recent decades, the introduction of machine learning and deep learning models has accelerated the classification of galaxies, leading to the replacement of manual classification methods with automated machine-based approaches. Recently, Vision Transformers (ViTs) have emerged as a significant innovation in machine learning, demonstrating substantial potential in various research fields. These models have particularly garnered attention in the analysis, detection, and classification of images and computer vision, due to their ability to process large datasets and learn complex patterns. The need to develop advanced methods for the automatic analysis of galaxy images to increase detection and classification accuracy in the shortest possible time motivated the current research to classify galaxy images from the Galaxy10 DECaLS dataset into 10 classes with an accuracy of 99.85% using the ViT model. The results obtained have been promising in comparison with other competitors.