Stone pine (Pinus pinea L.) is an important Mediterranean species known for its edible seeds, the pine nuts, the most expensive nuts in the world. In Chile, more than 5000 hectares of P. pinea have been planted since 2014, with a consequent increase in the demand for field data to guide its management. The use of images captured by unmanned aerial vehicles (UAVs) in forest plantation surveys has shown to be a feasible technological solution. This study aimed to validate the use of light detection and ranging (LiDAR) data, obtained using UAV, to quantify cone production in a 30-year-old stone pine plantation located in central Chile. Forest attributes were measured in all trees (n = 175). Semi-automatic tree segmentation was performed, and models were fitted based on LiDAR metrics, LiDAR-derived allometry, and field-measured allometry to estimate individual tree annual cone production (kg tree⁻¹) (CN) and individual tree historical cone production (kg tree⁻¹) (HCN). Overall accuracy of tree detection ranged from 93.8% to 100%, depending on crown structure. CN regressions performed similarly, showing values of R² from 0.41 to 0.43, RMSE from 1.44 kg to 1.97 kg, and rRMSE from 73.6% to 83.2%. For HCN, LiDAR-based regression performance was higher than that of field allometry (R² of 0.67 vs 0.44, RMSE of 18.93 kg vs 24.95 kg, and rRMSE of 42.5% vs 56.0%, respectively), showing correlation values of 0.82 and 0.74 for train and test data sets, respectively. Regardless of data origin, the best-performing regressions included variables related to tree height and crown area. LiDAR-based data acquisitions hold significant potential for stone pine management in intensive cone-production oriented plantations, supporting the need for accurate forest-structure data acquisition to improve cone yield models.