We use the ocean/atmosphere/biosphere box model of the global carbon cycle BICYCLE (Köhler et al., 2005) to reproduce low frequency changes in atmospheric CO2 as seen in Antarctic ice cores during the last glacial cycle (120,000 years) (Köhler et al., 2006). We force the model forward in time by various paleo-climatic records derived from ice and sediment cores. The simulation results of our proposed scenario match a compiled CO2 record from various ice cores with high accuracy (r2 = 0.89). The processes that contribute most to the glacial/interglacial changes in CO2 are variations in the sedimentation and dissolution rates of CaCO3, ocean circulation, ocean temperature and glacial iron fertilization of the marine biota in the Southern Ocean. The BICYCLE model includes also calculations for the carbon isotopes 13C and 14C and we assess what changes in atmospheric D14C might be based on variations in the carbon cycle. Our results suggest that during the last glacial cycle in general less than 120 o/oo of the increased atmospheric D14C are based on variations in the carbon cycle, while the largest part of the variations has to be explained by changing 14C production rates. Processes acting on the global carbon cycle that increase glacial D14C are a restricted glacial gas exchange between the atmosphere and the surface ocean through sea ice coverage, a reduced glacial ocean circulation, and the enrichment of DIC with 14C in the surface waters through isotopic fractionation during higher glacial marine export production caused by iron fertilization.From the available D14C data covering the last 50,000 years and our carbon cyclebased simulation results we can infer changes in the 14C production rates, which are then compared with two other estimates based on 10Be and geomagnetic field reconstruction. The agreements and discrepancies between these three independent approaches to estimate the 14C production rates are discussed and highlight the limitations and possible uncertainties in all three approaches.