In the current global warming context, rapid changes in Arctic sea ice have a significant impact on the environment, the hydrological conditions and ecosystems of the Arctic Ocean. As an important indicator of global climate change, accurate predictions of future changes in Arctic sea ice contribute to the understanding and awareness of climate change trends in the Arctic and even globally. At present, numerical models are one of the most important means to effectively estimate future changes in sea ice, but due to the many factors affecting the accuracy of the patterns, the estimated results between the models vary greatly. Therefore, this paper sorted out the current research progress on short term forecast and medium-long term prediction models of the Arctic Ocean at home and abroad, focusing on the simulation accuracy of the key elements in the Arctic sea ice such as the concentration, extent and area, thickness and drift speed of various models. It analyzed the prediction of the future change trend of Arctic sea ice in the current stage of the model. In addition, on the basis of previous research, the research and application of future numerical models in the prediction of Arctic Ocean sea ice are prospected.