Assessing information on climate change over a regional scale is made possible through general circulation models (GCMs). However, developers generally have a dilemma in selecting suitable GCM for regional scale downscaling to reduce the computational burden. Ranking of GCMs based on various conditions will help these purposes, and the present study evaluates the performance of GCMs using various performance evaluation parameters for ranking. Performance of twenty-six Coupled Model Intercomparison Project Phase 5 (CMIP5) GCMs was assessed in the present study to evaluate and rank the predictability of near-surface air temperature (tas) and precipitation (pr). The non-parametric trend existing in observed data from 35 stations is compared with GCM projected trends using Mann-Kendall trend analysis to assess the model reliability. Performance evaluation parameters such as percentage BIAS (PBIAS %), normalized root mean squared error (NRMSE %) and coefficient of determination (R2). Neither of the CMIP5 GCM performed consistently well throughout all four seasons. Also, models performing better in projecting temperature statistics are poor in capturing the precipitation trends and vice versa. The seasonal ranking of GCMs based on their ability to reproduce the regional weather condition would help in selecting suitable GCM for regional climate studies. © 2020, Saudi Society for Geosciences.