Abstract: The development strategy of regional integration in the Yangtze River Delta region has risen to a national strategy, which requires the establishment of interregional coordination mechanism to guide lowcarbon industries and make up for the shortcomings of green development. Mastering the spatiotemporal characteristics and influencing factors of carbon emissions efficiency is of great practical significance for the low carbon development and the realization of carbon emission reduction target in the Yangtze River Delta region. Based on the combination of SBM model and window analysis, the carbon emissions efficiency was calculated. The spatiotemporal characteristics of carbon emissions efficiency in the Yangtze River Delta from 1995 to 2017 was analyzed by traditional statistical analysis and spatial analysis. Meanwhile, the spatial Durbin panel model was constructed to analyze its influencing factors. The results are shown as follows: (1) The carbon emissions efficiency showed a fluctuating trend over the period of 1995-2017. In 2017, Shanghai, Suzhou and Wuxi had the largest carbon emissions efficiency of 1. The absolute difference and relative difference in carbon emissions efficiency showed fluctuating trend, and the change trend is basically consistent. There was an imbalance in the carbon emissions efficiency. (2) There was obvious spatial disparity for carbon emissions efficiency, the spatial pattern of carbon emissions efficiency changed little since 2005. The high carbon emissions efficiency areas were mainly concentrated in Shanghai, Suzhou and Wuxi. The gravity center of carbon emissions efficiency was mainly distributed in Nanjing City, and moved to the northwest. (3) The results of spatial Durbin panel  model show that improving the technical  level is an important way to change the carbon emissions efficiency. Urbanization and spatial factors played a significantly positive effect role in improving carbon emissions efficiency, foreign investment, energy consumption per unit GDP and ecological environment  have a significantly negative impact on carbon emissions efficiency.