This paper explores the application of quantum hybrid vision transformers for event classification in high-energy physics, specifically distinguishing photons and electrons in the electromagnetic calorimeter of the CMS experiment at the LHC. The authors construct several variations of quantum hybrid vision transformer models and compare their performance against classical vision transformer architectures. The quantum models aim to reduce computational resources while maintaining predictive power, addressing the computational challenges associated with large datasets and complex data. The study uses a dataset of 498,000 simulated electromagnetic shower events, with half originating from photons and the other half from electrons. The models are trained using a combination of classical and quantum components, including multi-head attention mechanisms. The results show that the quantum hybrid models achieve comparable performance to their classical counterparts with a similar number of parameters, suggesting potential advantages in terms of efficiency and scalability. The paper discusses the implications of these findings and outlines future directions for further research, including the exploration of more quantum elements in the models.This paper explores the application of quantum hybrid vision transformers for event classification in high-energy physics, specifically distinguishing photons and electrons in the electromagnetic calorimeter of the CMS experiment at the LHC. The authors construct several variations of quantum hybrid vision transformer models and compare their performance against classical vision transformer architectures. The quantum models aim to reduce computational resources while maintaining predictive power, addressing the computational challenges associated with large datasets and complex data. The study uses a dataset of 498,000 simulated electromagnetic shower events, with half originating from photons and the other half from electrons. The models are trained using a combination of classical and quantum components, including multi-head attention mechanisms. The results show that the quantum hybrid models achieve comparable performance to their classical counterparts with a similar number of parameters, suggesting potential advantages in terms of efficiency and scalability. The paper discusses the implications of these findings and outlines future directions for further research, including the exploration of more quantum elements in the models.