The difference in magnitude from this work (structural fitting) and the photometry catalog (aperture photometry) as a function of photometry catalog magnitude, for all galaxies in the SUPER_DUPER catalog in all 20 NIRCam medium or broad bands. We also show the population median and 1σ scatter in magnitude bins as error bars. Based on the median of these distributions, we find no significant systematic differences between the photometry catalog fluxes and pysersic recovered fluxes at m_AB < 26. However, structural fitting reports systematically higher fluxes than aperture photometry at m_AB > 26, especially in the broadband images, where the magnitude limit of our fitting selection is effectively fainter. These systematic differences in the faint regime are likely driven by the intrinsic difference between the Sérsic profile and the aperture loss correction, in terms of the assumption of the surface brightness curve of growth in the source outskirt.