Upcoming supernova surveys, such as the Nancy Grace Roman Space Telescope (Roman) and the Vera Rubin Observatory’s LSST, are expected to discover over a million supernovae, vastly increasing current samples. Using Type Ia Supernovae (SN Ia) to constrain cosmological parameters, including the dark energy equation-of-state parameters w₀ and wₐ, requires precise spectro-photometric models. One of the largest systematics in supernova cosmology analysis is the supernova SED model’s accuracy, which depends on survey photometric calibration. My team uses catalog-level simulations of LSST and Roman, incorporating forecasted calibration uncertainties, to assess their impact on model recovery, distance measurements, and inferred cosmology.