Land model benchmarking provides a comprehensive evaluation of land models that goes beyond traditional (“it looks better”) validation, with an established level of a priori performance for a number of metrics, for example, modeled surface fluxes within some “error” value of the observations or PDFs of observed and modeled surface fluxes overlapping by some threshold percentage, where metrics depend on a particular community, such as climate, weather, hydrology, carbon, etc. Land model (physics) should be able to out-perform, e.g. simple empirical models or climatology, where empirical approaches may be statistical, neural networks, or other machine learning where the relationship between training and testing data sets are manipulated to see how well a physically-based model utilizes available information. The Global Energy and Water Cycle Experiment (GEWEX), Global Land Atmosphere System Study (GLASS) and GEWEX Hydroclimatology Panel (GHP) have undertaken a joint project on land surface model benchmarking where the GLASS community provides benchmarking analysis tools, i.e. Protocol for the Analysis of Land Surface models (PALS; www.pals.unsw.edu.au), and GHP provides flux site data sets for different regions, seasons and variables, to evaluate e.g. energy, water and carbon budget components. We report on this effort.