One fundamental issue in managing bike sharing systems is the bike ow prediction. Due to the hardness of predicting the ow for a single station, recent research works o en predict the bike ow at cluster-level. While such studies gain satisfactory prediction accuracy, they cannot directly guide some ne-grained bike sharing system management issues at station-level. In this paper, we revisit the problem of the station-level bike ow prediction, aiming to boost the prediction accuracy leveraging the breakthroughs of deep learning techniques. We propose a new multi-graph convolutional neural network model to predict the bike ow at station-level, where the key novelty is viewing the bike sharing system from the graph perspective. More speci cally, we construct multiple interstation graphs for a bike sharing system. In each graph, nodes are stations, and edges are a certain type of relations between stations.en, multiple graphs are constructed to re ect heterogeneous relationships (e.g., distance, ride record correlation). A erward, we fuse the multiple graphs and then apply the convolutional layers on the fused graph to predict station-level future bike ow. In addition to the estimated bike ow value, our model also gives the prediction con dence interval so as to help the bike sharing system managers make decisions. Using New York City and Chicago bike sharing data for experiments, our model can outperform state-of-the-art station-level prediction models by reducing 25.1% and 17.0% of prediction error in New York City and Chicago, respectively.