Spectra and best-fit models for the WASP-107 b study. Red lines indicate the case where the NIRCam data is considered; blue lines indicate the case where the NIRSpec data is considered. We also show the contribution of the different components in the model to the best-fit spectrum for the NIRCam (FRECKLL) case. While the retrieval results are similar, systematic differences exist between the NIRCam and NIRSpec data for λ < 3.7 μm. This difference could be due to remaining instrumental systematics or a time-varying cloud cover. Note that the observations are plotted with the vertical offsets corrected.