Keeping the individual features and the complicated relations, graph data are widely utilized and investigated. Being able to capture the structural information by updating and aggregating nodes' representations, graph neural network (GNN) models are gaining popularity. In the financial context, the graph is constructed based on real-world data, which leads to complex graph structure and thus requires sophisticated methodology. In this work, we provide a comprehensive review of GNN models in recent financial context. We first categorize the commonly-used financial graphs and summarize the feature processing step for each node. Then we summarize the GNN methodology for each graph type, application in each area, and propose some potential research areas.