Currently, there are so many movie genres available to the general public, making it difficult for viewers to choose a movie. One of the most popular movies is the “Marvel Movies” or MCU (Marvel Cinematic Universe), which has become the highest grossing franchise of all time with 90 movies released. The large number of movie titles makes it difficult for people to choose which movie to watch. Therefore, a Marvel movie recommendation system is needed using a hybrid item-based and content-based filtering method. The content-based method calculates the similarity between movies by identifying similar Marvel movies based on content such as genre, actor, director, and synopsis. Meanwhile, item-based completes content-based recommendations by considering user preferences. The reason for using the hybrid item-based and content-based filtering method is to be able to produce more accurate recommendations than a single method. The types and sources of data used are secondary data from journals and the internet (Imdb and Movielens), as well as datasets about Marvel movies. From the results of testing the hybrid model, the precision value is 0.8 or 80% which indicates that the model is accurate. In item-based filtering testing, the similarity result of 0.68 shows good item similarity. In the content-based filtering test, the highest similarity is 0.14 and the lowest similarity is 0.10 which shows that the similarity between the searched content and the generated content is relevant.