South Africa is the leading soybean producer in Africa, contributing approximately 35% of the continent’s total production. Soybean<br />is important for national food security and agricultural sustainability, serving as a key nitrogen-fixing crop that supports soil fertility and economic growth. Monitoring biochemical parameters such as leaf chlorophyll content (LCC) is essential for assessing soybean health; however, cultivar-level variability can complicate the use of remote sensing-based approaches. This study evaluates the performance of four machine-learning algorithms, eXtreme Gradient Boosting (XGBoost), Random Forest (RF), Partial Least Squares Regression (PLSR), and Artificial Neural Network (ANN), using unmanned aerial vehicle (UAV)-based data across 15 soybean cultivars during the early reproductive phase. Results show that model performance is strongly cultivar dependent. Tree-based models achieved the highest accuracy, with XGBoost and RF reaching Root Mean Square Error (RMSE) values as low as 2.9 μmol m⁻² for PHIP62T16R and R² values up to 0.96 for RA655R. In contrast, ANN and PLSR performed substantially worse for cultivars with more complex spectral responses, such as PAN1555R. Residual analyses from generalised models revealed systematic over- and underprediction in several cultivars, indicating that pooled models could not fully account for cultivar-specific spectral differences. Variable importance analyses identified red-edge, near-infrared (NIR), and greenness-enhancing indices as the most influential predictors of LCC. Overall, the study demonstrates that incorporating cultivar information and using stratified model calibration significantly improves the reliability of UAV-based chlorophyll monitoring in heterogeneous soybean canopies.