Improving theoretical tools for materials that contain transition-metal and lanthanide atoms drives this research in computational physics. Standard colinear density functional theory (DFT) methods often struggle to describe the complex magnetic behavior of these systems, particularly for lanthanides. The project evaluates recent advances in non-colinear two-component density-functional approaches to determine whether they provide a more accurate picture of molecular magnetic qubits. Work proceeds in 2 stages. Initial calculations examine single-center metallocenes, followed by full treatment of the Mn12-Acetate molecule, to test whether non-colinear magnetic anisotropy explains higher-order anisotropies observed experimentally. The second stage investigates how self-interaction errors affect predictions for f-electron molecular magnets. The effort supports continued fellowship training for Dr. Winfred Mueni Mulwa in computational physics while advancing methods essential for quantum information and molecular magnet design.