Climate change is significantly affecting agriculture and food security. Variations in temperature and uncertainties in rainfall patterns negatively affect the yield of crops. Crop models have been widely used in assessing the impact of climate change on agricultural crops. The study was planned to evaluate the potential impacts of climate change on wheat yield in different agro-ecological zones of Khyber Pakhtunkhwa (KPK), Pakistan. The CERES-Wheat model was calibrated and evaluated with the experimental dataset. The survey data of 50 farms were collected from the study districts Chitral, Dera Ismail (D.I.) Khan, and Peshawar. Climate change scenarios were generated on the basis of five selected general circulation models (GCMs) under representative concentration pathways (RCPs 4.5 and 8.5) for mid-century (2040–2069). Model was calibrated well and showed a close agreement between observed and simulated values of grain yield with root mean square error (RMSE) ranging from 25 to 55 kg ha−1 for three cultivars, i.e., Faisalabad-2008, Galaxy-2013, and PirSabak-2013. Climate projections showed that there would be increases in Tmax of 2.33 °C and 2.93 °C and Tmin of 2.16 °C and 3.13 °C under RCPs 4.5 and 8.5 respectively. Climate change impacts showed that wheat yield will increase by 10% in RCP 4.5 and 14% in RCP 8.5 at Chitral, while yield will be reduced by 7.9% in RCP 4.5 and 11% in RCP 8.5 in D.I. Khan. The yield will also be reduced at Peshawar by 5.5% in RCP 4.5 and 8.4% in RCP 8.5.