 The study compares the performance of the UK climate projections 2018-UKCP-18, with observational and re-analysis data to quantify model accuracy in capturing mean, extremes, and variability in the climate state and heat stress metrics. The results show that UKCP-18 simulations generally perform better than CMIP-5 models in reproducing observed spatial patterns of UK climate related to extreme heat, with a minor improvement in model performance achieved by increasing spatial resolution. The study also finds that model error in capturing characteristics of extreme heat is reduced when using heat stress metrics with a larger vapor pressure component. Finally, the study shows that the 95th percentile of observed UK summer temperature has greater inter-annual variability than the summer mean over the recent past, which is underestimated by UKCP-18 models compared to had UK grid and ERA-5. The study also finds that UK summer mean and 95th percentile temperatures are projected to increase at a faster rate than the global mean temperature in the future. This article was authored by A.T. Kennedy Arthur, O. Andrews, D. M. Mitchell, and others.