China proposed a 2060 carbon neutrality target in 2020 and proposed to accelerate the implementation of the 2030 national carbon peaking mandate. Considering the important mission and great potential of cities in national carbon reduction efforts, as well as the significant differences in their carbon emissions in terms of total volume, structure and trends, an in-depth understanding of the typological characteristics of China's urban carbon peaking trends is important for local governments to design and carry out differentiated peaking actions. In this paper, the static and dynamic factors influencing urban peak carbon trends are considered, and the K-means clustering algorithm is used to classify and analyze the peak trends of 286 sample cities in China. The results show that Chinese cities can be classified into five categories. Finally, the paper makes practical suggestions on the design of targets and action priorities for urban carbon peaking for different types of cities.