The arrival of the Internet of Things (IoT) era has accelerated the mutual integration of various industries and IoT technologies while developing its influence to a broader and more profound level. IoT based on Graph Neural Networks is a research direction that applies Graph Neural Networks to the field of IoT. IoT is a network of various smart devices, sensors, and communication technologies that can interact and communicate with each other to form a vast and dynamically changing network. Graph Neural Network is a special neural network structure that can handle unstructured data and model complex data relationships and dependencies. Traditional machine learning algorithms are not able to capture well the complex correlations that exist between devices in IoT. Still, Graph Neural Networks can model and analyze IoT data by constructing connectivity relationships between devices. IoT research based on Graph Neural Networks is a new field emerging in recent years, which combines Graph Neural Networks and IoT to provide a new idea and method for data modeling and analysis in the IoT field. This article focuses on common Graph Neural Networks (GNNs) and IoT, as well as applications of GNNs in the field of IoT. Meanwhile, it looks into the future development trend of IoT based on Graph Neural Networks.