Comparison between the fluxes measured from the PRISM and G395M spectra using the [O III]λλ4960, 5008 emission line doublet. The blue solid line shows the running median. We found a systematic offset between the two gratings despite the fact that the slit losses and spectral extraction method used are identical. The PRISM fluxes are brighter by approximately 10−15%. This offset does not depend significantly on the line flux itself and likely points to a calibration issue.