Significant progress in permafrost carbon science made over the past decades include the identification of vast permafrost carbon stocks, the development of new pan‐Arctic permafrost maps, an increase in terrestrial measurement sites for CO<jats:sub>2</jats:sub> and methane fluxes, and important factors affecting carbon cycling, including vegetation changes, periods of soil freezing and thawing, wildfire, and other disturbance events. Process‐based modeling studies now include key elements of permafrost carbon cycling and advances in statistical modeling and inverse modeling enhance understanding of permafrost region C budgets. By combining existing data syntheses and model outputs, the permafrost region is likely a wetland methane source and small terrestrial ecosystem CO<jats:sub>2</jats:sub> sink with lower net CO<jats:sub>2</jats:sub> uptake toward higher latitudes, excluding wildfire emissions. For 2002–2014, the strongest CO<jats:sub>2</jats:sub> sink was located in western Canada (median: −52 g C m<jats:sup>−2</jats:sup> y<jats:sup>−1</jats:sup>) and smallest sinks in Alaska, Canadian tundra, and Siberian tundra (medians: −5 to −9 g C m<jats:sup>−2</jats:sup> y<jats:sup>−1</jats:sup>). Eurasian regions had the largest median wetland methane fluxes (16–18 g CH<jats:sub>4</jats:sub> m<jats:sup>−2</jats:sup> y<jats:sup>−1</jats:sup>). Quantifying the regional scale carbon balance remains challenging because of high spatial and temporal variability and relatively low density of observations. More accurate permafrost region carbon fluxes require: (a) the development of better maps characterizing wetlands and dynamics of vegetation and disturbances, including abrupt permafrost thaw; (b) the establishment of new year‐round CO<jats:sub>2</jats:sub> and methane flux sites in underrepresented areas; and (c) improved models that better represent important permafrost carbon cycle dynamics, including non‐growing season emissions and disturbance effects.</jats:p>