<p>The preparation of cutting tool edges is essential for machining efficiency and can be enhanced through drag finishing, which rounds the edge using abrasive media to improve precision and tool life. This study evaluates the performance of boosting algorithms, LightGBM and CatBoost, in predicting the wear of tungsten carbide end mills treated with two different abrasive media and without preparation, under varying cutting parameters (axial depth, radial depth, and cutting speed) while machining AISI 4140 steel. The results demonstrate strong predictive capabilities for both models, with <i>R</i>² values of 0.9480 for CatBoost and 0.9418 for LightGBM, the latter showing a 47% reduction in training time. The techniques employed in this study explore the relationship between the cutting edge condition and wear and prove to be a suitable tool for analyzing machining with different cutting tools.</p>