Abstract Recent controversy regarding the existence of massive ( log ( M * / M ⊙ ) ≳ 11 ) galaxies at z > 6 poses a challenge for galaxy formation theories. Hence, it is of critical importance to understand the effects of SED fitting methods on stellar mass estimates of Epoch of Reionization galaxies. In this work, we perform a case study on the AGN host galaxy candidate COS-87259, with spectroscopic redshift z spec = 6.853, that is claimed to have an extremely high stellar mass of log ( M * / M ⊙ ) ∼ 11.2 . We test a suite of different SED fitting algorithms and stellar population models on our independently measured photometry in 17 broad bands for this source. Between five different code setups, the stellar mass estimates for COS-87259 span log ( M * / M ⊙ ) = 10.24 –11.00, while the reduced χ 2 values of the fits are all close to unity within Δ χ ν 2 = 1.2 , such that the quality of the SED fits is basically indistinguishable. Only when we adopt a nonparametric star formation history model within Prospector do we retrieve a stellar mass exceeding log ( M * / M ⊙ ) = 11 . Although the derived stellar masses change when using previously reported photometry for this source, the nonparametric SED-fitting method always yields the highest values. As these models are becoming increasingly popular for James Webb Space Telescope high-redshift science, we stress the absolute importance of testing various SED fitting routines particularly on apparently very massive galaxies at such high redshifts.