Based on land use, night-time light, POI data, etc., we used the coupling coordination degree and XGBoost-SHAP model, the response characteristics of ecosystem service value (ESV) and the spatial differentiation and driving factors of coupling coordination degree (CCD) under the multi-dimensional urban-rural gradient of Chengdu-Chongqing Economic Circle from 2012 to 2022 were analyzed from the grid scale. It was found that the ESV of the Chengdu-Chongqing Economic Circle showed an increasing trend during the study period, with a total increase of 12.98 billion CNY. In terms of spatial differentiation along the urban-rural gradient, the proportion of suburban space was observed to rise from 0.55% to 0.89%, representing a growth rate of 61.8%, making it the fastest-growing spatial category and indicating a notable trend of suburbanization. The proportion of urban space increased from 0.15% to 0.21%, while that of rural space decreased from 96.67% to 94.25%. A negative correlation between CCD and ESV was identified in the urbanization process of the Chengdu-Chongqing Economic Circle. The CCD was found to be at the preliminary coordination stage and exhibited a slight declining trend. Along the transition from rural to urban areas, ESV was shown to decrease, while CCD demonstrated an increasing trend. Key factors influencing the urban-rural gradient and the CCD of ecosystem service value were identified as population density, normalized difference vegetation index (NDVI), PM_2.5 concentration, and land use change, all of which were revealed to exhibit nonlinear effects.