This paper introduces a hybrid quantum vision transformer (QViT) for event classification in high-energy physics, specifically for distinguishing photons and electrons in the electromagnetic calorimeter (ECAL) of the CMS detector. The model is compared against classical vision transformers (ViT) to evaluate its performance. The hybrid model uses a combination of quantum and classical components, with the quantum part implementing a hybrid multi-head attention mechanism. The model is trained on a dataset of simulated electromagnetic shower events, with each event represented as an image grid containing energy and timing information. The model's performance is evaluated on a validation and test set, with results showing that the hybrid model achieves comparable performance to the classical model with a similar number of parameters. The study highlights the potential of quantum machine learning to offer computational advantages in high-energy physics applications, where data complexity and computational resources are significant challenges. The hybrid model's architecture is described, along with its training process and hyperparameters. The results indicate that the hybrid model can be a viable alternative to classical models in high-energy physics, particularly in scenarios where computational efficiency is crucial. The study also discusses the limitations of the current quantum model and suggests future research directions to improve its performance.This paper introduces a hybrid quantum vision transformer (QViT) for event classification in high-energy physics, specifically for distinguishing photons and electrons in the electromagnetic calorimeter (ECAL) of the CMS detector. The model is compared against classical vision transformers (ViT) to evaluate its performance. The hybrid model uses a combination of quantum and classical components, with the quantum part implementing a hybrid multi-head attention mechanism. The model is trained on a dataset of simulated electromagnetic shower events, with each event represented as an image grid containing energy and timing information. The model's performance is evaluated on a validation and test set, with results showing that the hybrid model achieves comparable performance to the classical model with a similar number of parameters. The study highlights the potential of quantum machine learning to offer computational advantages in high-energy physics applications, where data complexity and computational resources are significant challenges. The hybrid model's architecture is described, along with its training process and hyperparameters. The results indicate that the hybrid model can be a viable alternative to classical models in high-energy physics, particularly in scenarios where computational efficiency is crucial. The study also discusses the limitations of the current quantum model and suggests future research directions to improve its performance.