The present report outlines a workflow integrating various virtual screening methods to identify potential antimalarial compounds. To develop QSAR models, a dataset of 2,314 compounds was analyzed using linear discriminant analysis and the QuBiLs-MAS software. 37 individual models were generated and subsequently combined into a fusion-based multiclassifier system (MCS), which achieved predictive performances of 91.35% for the training set and 92.06% for the test set. The MCS was further evaluated through a virtual screening simulation involving 13,410 compounds from GlaxoSmithKline, yielding an extrapolation rate of 91.43%. Following this, several drug-likeness filters, the finalized MCS, and chemical diversity analyses were applied to select candidate compounds from three datasets for parasitological assays. Using the proposed in silico pipeline, a total of 6,811 drugs, 15,000 chemical compounds, and 1,120 biologically active molecules from the DrugBank, PrintScreen15, and Tocriscreen collections, respectively, were virtually screened. From these, 80 compounds were shortlisted as potential antimalarial candidates. Ultimately, 15 compounds were purchased and tested in vitro against two Plasmodium falciparum strains (3D7 and Dd2). Of these, five drugs (ziprasidone, isradipine, amcinonide, triflupromazine, and anisotropine) and four chemical compounds (NGB 2904, A23187, Otava-7019050991, and Otava-1677649) demonstrated antimalarial activity, with \(\text {IC}_{50}\) values \(\le 15 \upmu \text {M}\) . This approach represents a promising computational tool for the early stages of antimalarial drug discovery.