Liquid crystals exhibit an exceptionally rich array of topological defect structures and multistable characteristics under geometric confinement or varying thermodynamic conditions. Understanding and precisely manipulating these stable states, alongside their transition pathways, is of fundamental importance for both theoretical physics and designing novel optical metamaterials. However, the highly nonlinear free energy functionals governing liquid crystals (such as the Landau-de Gennes model) pose significant computational challenges in identifying all stable configurations. This talk will systematically explore the application and recent breakthroughs of the "solution landscape" framework in unveiling the complex multistability of liquid crystals. Using high-index saddle dynamics method, we not only captures all local minima (stable states) but also locates the saddle points and delineates their intricate connectivity network. Ultimately, the solution landscape provides a rigorous mathematical description of exotic NLC defects, offering powerful predictive guidance for designing intelligent, multistable soft materials.