<p>Vehicle-assisted edge caching is a promising approach to alleviate the pressure on the core network. Vehicles cache some of the data in advance and share content with users via vehicle-to-everything (V2X), improving the performance and user experience of the Internet of Vehicles (IoV). However, in the 6&#xa0;G era, the limited computing and storage capacity of vehicles makes it difficult to meet the demand for content caching, and the high demand for real-time performance makes the importance of content freshness increasingly prominent. At this point, the choice of caching strategy is crucial. In this paper, we propose a mobile edge caching strategy based on content freshness and user preference for V2X content-sharing scenarios in IoV. First, we construct a device mobility model based on long short-term memory (LSTM) and a user preference prediction model based on social proximity, content theme classification, and user historical request records. In addition, under the constraints of the vehicle’s and the user’s cache capacities, we propose an optimization problem aiming to maximize the average freshness and obtain a sub-optimal solution to the optimization problem using a user preference prediction-based greedy algorithm. Simulation results show that the optimization algorithm proposed in our paper predicts users’ interests more accurately, reduces the transfer latency and energy consumption ratio effectively, and maximizes the average freshness of the IoV.</p>