In this study, a simple machine learning-based quantitative structure-activity relationship (QSAR) model has been developed to predict the inhibitory potency (pIC50 values) of FLT3 tyrosine kinase inhibitors, crucial for the treatment of acute myeloid leukemia. Distinctively, our model leverages an extensive and diverse dataset, 14 times larger than those employed in prior studies within this field, enabling an unparalleled scope of compound analysis. This vast dataset, combined with further exploration of molecular descriptors, allows for more accurate and efficient predictions than previous studies. The Random Forest Regressor algorithm, selected for its excellent predictive performance, was trained on a dataset comprising 1080 compounds and subjected to rigorous validation. Internal validation, performed through Leave-One-Out (LOO) cross-validation, yielded an Q2LOO value of 0.92, demonstrating the model's internal consistency and predictive reliability. Additionally, external validation using 270 independent compounds not included in the training set achieved an R2 of 0.941, with a standard deviation of 0.237, outperforming previous approaches in terms of simplicity and precision. Key molecular descriptors have been identified, enhancing the understanding of structural requirements for inhibitor potency. Additionally, a user-friendly computational tool has been developed, enabling rapid predictions of pIC50 values. Using this tool, potential FLT3 inhibitors have been identified via ligand-based virtual screening. This study represents a significant advancement in FLT3 inhibitor discovery using a simplified QSAR-machine learning model, enabling more precise and practical identification of potential drug candidates at an early stage, promising faster development of targeted therapies and streamlining the ligand-based drug design process.