Recent studies indicated that Arctic lakes play an important role in receiving, processing, and storing organic carbon exported from terrestrial ecosystems. To quantify the contribution of Arctic lakes to the global carbon cycle, we developed a one‐dimensional process‐based Arctic Lake Biogeochemistry Model (ALBM) that explicitly simulates the dynamics of organic and inorganic carbon in Arctic lakes. By realistically modeling water mixing, carbon biogeochemistry, and permafrost carbon loading, the model can reproduce the seasonal variability of CO2 fluxes from the study Arctic lakes. The simulated area‐weighted CO2 fluxes from yedoma thermokarst lakes, nonyedoma thermokarst lakes, and glacial lakes are 29.5, 13.0, and 21.4 g C m−2 yr−1, respectively, close to the observed values (31.2, 17.2, and 16.5 ± 7.7 g C m−2 yr−1, respectively). The simulations show that the high CO2 fluxes from yedoma thermokarst lakes are stimulated by the biomineralization of mobilized labile organic carbon from thawing yedoma permafrost. The simulations also imply that the relative contribution of glacial lakes to the global carbon cycle could be the largest because of their much larger surface area and high biomineralization and carbon loading. According to the model, sunlight‐induced organic carbon degradation is more important for shallow nonyedoma thermokarst lakes but its overall contribution to the global carbon cycle could be limited. Overall, the ALBM can simulate the whole‐lake carbon balance of Arctic lakes, a difficult task for field and laboratory experiments and other biogeochemistry models.