Abstract: Promoting the green and low-carbon transformation of the economy is the inevitable requirement of implementing Xi’s ecological civilization thought in the new era. Green technology innovation is the core support to achieve the “double carbon” goal, but the academic community is still lack of in-depth discussion on the impact of green technology innovation on carbon emissions and its spatial effect in developing countries.Based on the panel data of 110 cities in the Yangtze River economic belt from 2006 to 2020, the spatial effect of green technology innovation on carbon emissions is studied by using GIS spatial analysis and spatial econometric model. The results show that: (1) from 2006 to 2020, the spatial distribution of green technology innovation and carbon emissions in the cities of the Yangtze River economic belt is unbalanced, and the spatial polarization of green technology innovation is significant. The innovation pattern dominated by the core cities in the Yangtze River Delta and promoted by the capital cities in the central and western regions is increasingly prominent, and the high-carbon regions are distributed in the downstream regions and scattered in the upper and middle regions; There is a significant positive spatial correlation between green technology innovation and carbon emissions, and the evolution of spatial agglomeration shows spatial lock-in and path dependence. (2) The impact of green technology innovation on carbon emissions shows a significant “U” shaped relationship, and the spatial spillover effect shows an inverted “U” shaped opposite relationship, reflecting the “rebound effect” and “polarization trickle down” effect of technological innovation. (3) Heterogeneity analysis shows that compared with resource-based cities, the carbon emission reduction effect of green technology innovation in non resource-based cities is stronger, while the carbon emission increase effect is weaker; The impact of green technology innovation in upstream cities on carbon emissions is mainly reflected in the short term, with the direct effect of “U” type and the indirect effect of “inverted U” type. The impact of green technology innovation in midstream cities is mainly reflected in the long term, with the direct effect and indirect effect of “inverted U” type. The impact of downstream cities is not significant in both the short term and the long term.