The primary focus of this research is the development and application of cost efficient multirefer-ence methods to quantitatively model chemical phenomena in excited states. Multi-configuration pair-density functional theory (MC-PDFT) is a method developed in the Gagliardi and Truhlar groups that aims at combining the advantages of wave function and density functional theories to allow robust modeling of strongly correlated systems. While the energies of MC-PDFT with a state-average reference have shown remarkable accuracy for reaction barriers and excitation energies, analytic gradients are necessary for additional applications like efficient calculations of stationary points on potential energy surfaces and direct dynamics simulations. The analytic gradients for MC-PDFT with a state-average reference wave function and density fitting were implemented in the OpenMolcas and PySCF software. Including density fitting made the code more performant and allowed for the study of significantly larger systems. This was then extended by formulating and implementing gradients and energies for compressed multi-state pair- density functional theory (CMS-PDFT). CMS-PDFT includes state interactions by allowing states to mix through the diagonalization of an effective Hamiltonian. As a result, this method gives the correct shape of the potential energy surface for systems with nearly degenerate states, e.g., near conical intersections or locally avoided crossings. In this work, we are interested in photosensitization, which is a process where a light-harvesting molecule transfers energy to a substrate and promotes it to an excited state. This excitation mech- anism is important for the photochemistry of Ir-Ni dual catalysis and photolesion formation in thio-substituted DNA. In both studies, the insights provided by the MC-PDFT calculations helped to determine the mechanism responsible for these photochemical reactions.