Recommendation systems enhance user’s experience and provides a solution to the issue of information overload effectively. In the big data era, many new models have been proposed and also achieved good results by combining recommendation algorithms with deep learning models. This paper analyzes the research status and achievements of combining deep learning models and recommendation algorithms. The advantages and the shortcomings of these deep learning recommendation model algorithms are analyzed. Third, the future development direction of the deep learning recommendation model is discussed. In addition, it discusses the future developing trend of deep learning recommendation models based on commonality of reviewed models and developing needs of recommendation systems.