Abstract. The temperature at the Antarctic ice sheet bed and the temperature gradient in subglacial rocks have been directly measured only a few times, although extensive thermodynamic modelling has been used to estimate geothermal heat flux under ice sheet. During the last five decades, deep ice-core drilling projects at six sites – Byrd, WAIS Divide, Dome C, Kohnen, Dome F, and Vostok – have succeeded in reaching to, or nearly to, the bed in inland locations in Antarctica. When temperature profiles in these boreholes and heat flow model are combined with estimations of vertical velocity, the heat flow at ice sheet base is translated to a geothermal heat flux of 117.8 ± 3.3 mW m−2 at Byrd, 67.3 ± 8.6 mW m−2 at Dome C, 79.0 ± 5.0 mW m−2 at Dome F, and −3.3 ± 5.6 mW m−2 at Vostok, close to predicted values. However, estimations at Kohnen and WAIS Divide gave flux of 161.5 ± 10.2 mW m−2 and 251.3 ± 24.1 mW m−2, respectively, far higher than that predicted by existing heat flow models. The question arises as to whether this high heat flow represents regional values, or if the Kohnen and WAIS Divide boreholes were drilled over local hot spots.