Better representations of biogeochemistry–climate feedbacks and ecosystem processes in Earth system models (ESMs) are essential for reducing uncertainties associated with projections of climate change. Model–data intercomparisons are required to inform the improvement of land carbon cycle models, and such efforts can inform the design of new measurement campaigns aimed at reducing uncertainties associated with key terrestrial processes. As ESMs become increasing complex, there is a growing need for comprehensive and multi-faceted evaluation and benchmarking of model results. The International Land Model Benchmarking (ILAMB) Package was designed to facilitate systematic and comprehensive model–data intercomparison and improve understanding of factors influencing model fidelity. We used ILAMB to benchmark land carbon cycle models coupled within ESMs used to conduct historical simulations for the Fifth and Sixth Phases of the Coupled Model Intercomparison Project (CMIP5 and CMIP6). Results published in the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report indicate that the suite of CMIP6 land models exhibit better performance than the suite of corresponding CMIP5 models in comparison with observations for a variety of biogeochemical, hydrological, and energy-related variables. These results provide a baseline from which improved benchmarks, informed by a growing collection of observational data, may help further constrain terrestrial carbon cycle processes and therefore projected climate change in future CMIP activities that employ emissions-forced simulations. For CMIP7, new evaluation challenges arising from higher resolutions and enhanced model complexity need to be rigorously addressed. The CMIP Climate Model Benchmarking Task Team aims to provide a vision and concrete guidance for establishing a systematic and rapid performance assessment of the large number of models expected to participate in CMIP7, across all facets of the models. This presentation will describe the results of the CMIP6 versus CMIP5 comparison for land models, as well as discussing progress on developments for CMIP7 analysis and benchmarking.