Abstract Wheat is one of the most important cereal crops in Mexico, but the impact of future climate change on production is not known. To quantify the impact of future climate change together with its uncertainty, two wheat crop models were executed in parallel, using two scaling methods, five Global Climate Models (GCMs) and two main Representative Concentration Pathways (RCPs) for the 2050s. Simulated outputs varied among crop models, scaling methods, GCMs, and RCPs; however, they all projected a general decline in wheat yields by the 2050s. Despite the growth-stimulating effect of elevated CO2 concentrations, consistent yield declines were simulated across most of the main wheat growing regions of Mexico due to the projected increase in temperature. Exceptions occurred in some cooler areas, where temperature improved sub-optimal conditions, and in a few areas where rainfall increased, but these increases only provided negligible contributions to national production. Larger and more variable yield declines were projected for rainfed wheat due to current and projected spatial variability of temperature and rainfall patterns. Rainfed wheat, however, only contributes about 6% of Mexico’s wheat production. When aggregating the simulated climate change impacts, considering temperature increase, rainfall change, and elevated atmospheric CO2 concentrations for irrigated and rainfed wheat cropping systems, national wheat production for Mexico is projected to decline between 6.9% for RCP 4.5 and 7.9% for RCP 8.5. Model uncertainty (combined for crop and climate models) in simulated yield changes, and across two scaling methods, was smaller than temporal and spatial variability in both RCPs. Spatial variability tends to be the largest in both future scenarios. To maintain or increase future wheat production in Mexico, adaptation strategies, particularly to increasing temperatures affecting irrigated wheat, or expanding the cropping area, will be necessary.