A radiometric field intercomparison was conducted at the Acqua Alta Oceanographic Tower (AAOT) in the Adriatic Sea from 14 to 21 July 2022 to assess differences in the accuracy of above-water radiometer systems (Sea-Bird HyperSAS, pySAS, TriOS-RAMSES) processed using an open-source community processor (HyperCP). Class-based and sensor-specific characteristics of the radiometers were used to determine the quantities Ed(λ), Lsky(λ), Lt(λ) and Rrs(λ), and their associated uncertainties. Using sensor-specific characteristics, the differences among systems were 2% for Ed(λ), Lsky(λ), Lt(λ) and 2.5% for Rrs(λ) and the uncertainties were 1.5%, 2%, 1.5% and 5%, respectively. The differences were higher when using class-based characteristics. Lwn(λ) values were also compared to the above-water AERONET-OC SeaPRiSM and in-water HyperPro II. For SeaPRiSM, the differences and uncertainties were <5% over blue and green bands when using Mobley [App. Opt.38, 7442 (1999)10.1364/AO.38.007442] sea surface reflectance factors (ρ) and no NIR correction. For HyperPro II the differences were larger but were reduced when Rrs(λ) from the above-water systems was computed using ρ from Zhang et al. [Opt. Express25, A1 (2017)10.1364/OE.25.0000A1] with similarity (sim) spectrum (spec) NIR correction. HyperCP using sensor-specific characteristics is highly recommended to reduce the associated uncertainties and to produce the highest quality data for satellite OC validation.