Variational autoencoders were proven successful in domains such as computer vision and speech processing. Their adoption for modeling user preferences is still unexplored, although recently it is starting to gain attention. In this paper, we propose a novel approach to collaborative filtering using sequential variational autoencoders (SVAEs). We demonstrate the effectiveness of SVAEs in modeling user preferences and improving the accuracy of recommendation systems. Our results show that SVAEs outperform traditional collaborative filtering methods and provide a more accurate and personalized recommendation experience for users.