Optimizing the structure of land use is essential to the low-carbon sustainable development of a region. This article takes Chengdu, a typical western China city, as the case study. First, carbon emission coefficients of land use are used to calculate the carbon emissions. Then, based on multi-objective linear programming (MOP), economic development priority scenario (S1), low-carbon economic development scenario (S2), and strengthening low-carbon economic scenario (S3) are proposed. Finally, the future land use simulation (FLUS) model is used to predict the spatial layout of land use under the three scenarios. The result shows that from 1990 to 2020, the carbon emissions increased by 7,617.61 thousand tons, with an annual growth rate of 3.75%. The main difference among the three scenarios is the occupied degree of farmland caused by the expansion of construction land, and the potential carbon reduction is 969.72 (5.2%), 2414.31 (13.1%), and 3878.89 tons (21.0%) in S1, S2, and S3, respectively. The FLUS model shows that conversion mainly occurs around the urban built-up area of Chengdu. This research can provide planning suggestions for the low-carbon development of Chengdu and a reference for other regions.