The conservation of the genetic diversity of native maize in Mexico is a priority due to its cultural, agricultural, and environmental importance. This study presents the development and evaluation of the imanr package, a computational tool based on Boosted Ensembles designed to automate the classification of racial complexes of native maize. Using a national database, a model was implemented that leverages morphological and geographical variables to provide precise and rapid classifications. The methodology included the optimization of key parameters through cross-validation, achieving up to 90% in balanced accuracy and a Cohen's Kappa coefficient of 0.84. These results highlight the robustness of the model compared to traditional methods, which rely on subjective expert judgment and require extended evaluation times. The findings demonstrate that the package not only surpasses conventional methods in terms of efficiency but also offers an accessible tool for conserving and monitoring native maize diversity, aligning with the recommendations of the Global Maize Project (PGMN). Moreover, its usability was enhanced by developing a graphical user interface, allowing non-specialized users to fully utilize its potential. imanr represents a significant advancement in native maize conservation science, contributing to the modernization of identification processes and strengthening sustainable management strategies for this essential genetic resource. This model directly addresses the need for innovative tools to monitor and preserve maize diversity in Mexico and suggests a promising pathway for future applications in the agricultural sector.