This study investigates the performance of four individual machine learning models (CatBoost, LightGBM, Random Forest (RF), and XGBoost) and a hybrid model for estimating forest above-ground biomass (AGB) in two study regions, east Jilin (JL) and central Guangxi (GX). The study evaluates the impact of forest types, independent variables, and spatial autocorrelation on AGB estimation accuracy. The hybrid model, which integrates the strengths of the individual models, consistently outperforms them in all scenarios. The results show that the RF model performs best in scenarios 5, 6, and 7, while the CatBoost model performs best in the remaining scenarios. The hybrid model, developed using an ensemble strategy, significantly improves estimation accuracy and stability, effectively addressing the challenge of model selection in forest AGB forecasting. The study also highlights the importance of spatial cross-validation in reducing the impact of spatial autocorrelation on AGB estimation. The findings suggest that the hybrid model provides a more accurate and stable approach for forest AGB estimation compared to individual models. The study concludes that the hybrid model is a promising approach for improving forest AGB estimation accuracy in different scenarios.This study investigates the performance of four individual machine learning models (CatBoost, LightGBM, Random Forest (RF), and XGBoost) and a hybrid model for estimating forest above-ground biomass (AGB) in two study regions, east Jilin (JL) and central Guangxi (GX). The study evaluates the impact of forest types, independent variables, and spatial autocorrelation on AGB estimation accuracy. The hybrid model, which integrates the strengths of the individual models, consistently outperforms them in all scenarios. The results show that the RF model performs best in scenarios 5, 6, and 7, while the CatBoost model performs best in the remaining scenarios. The hybrid model, developed using an ensemble strategy, significantly improves estimation accuracy and stability, effectively addressing the challenge of model selection in forest AGB forecasting. The study also highlights the importance of spatial cross-validation in reducing the impact of spatial autocorrelation on AGB estimation. The findings suggest that the hybrid model provides a more accurate and stable approach for forest AGB estimation compared to individual models. The study concludes that the hybrid model is a promising approach for improving forest AGB estimation accuracy in different scenarios.