Gathering the necessary information from the large amount of data available on the internet can be difficult because of its volume. To make it easier, we have recommender systems (RSs) that can generate suggestions or a path to follow. The RSs are categorized into various lines of action such as collaborative filtering, content-based, knowledge-based, and hybrid recommender systems. This paper centers on leveraging collaborative filtering (CF) and Vision Transformer (ViT) methodologies to develop an advanced RS, enhancing personalized content discovery for users. The CF utilizes the collective preferences of a user community to generate recommendations, effectively identifying patterns and predicting user interests. By integrating Vision Transformers, which excel in understanding and processing visual data, the proposed RS can offer more accurate and contextually relevant image-based recommendations. Vision Transformers provides a deep learning architecture that captures intricate features in images, enabling the system to comprehend user preferences at a granular level. The RS enhances the platform’s value and economic viability by consistently aligning with user preferences, boosting satisfaction, and fostering sustained business growth. Through the amalgamation of CF and ViT, our approach not only improves recommendation accuracy but also scales efficiently with growing data volumes because of item-based collaborative filtering (IBCF). This synergy between deep learning (DL) techniques and recommendation algorithms signifies a notable advancement in personalized content delivery, promising to elevate user engagement and retention. The proposed RS is evaluated on Amazon dataset with MAE, MSE, and RMSE evaluation measures which shows the effectiveness of the proposed approach.