The FeO content of sinter is an important index to measure the reducibility of sinter. In order to comprehensively and systematically outline the prediction methods of FeO content in various sinters，two prediction models of mechanism model and machine learning algorithm ( including classic single learning algorithm model，collaborative optimization algorithm model，deep neural network algorithm model and integrated learning algorithm model) are started to discuss from parameters based on the process，prediction recognition of the tail image，and prediction bused on imageparameter collaborative prediction. The development of the method for predicting FeO content in sinter is summarized; the emerging algorithm prediction models in recent years are comprehensively summarized and analyzed and the advantages and limitations of each algorithm in the sinter FeO content prediction model are expounded. At the same time，the future research direction of sinter FeO content prediction is prospected，in order to provide a useful reference for the research and practice in this field.