The term “integrated assessment model” (IAM) has been used since the 1980s to describe a class of models that today are focused on global-scale, long-term analyses of multi-system interactions among climate, land, energy, water, and socioeconomic systems. Examples include AIM, COFFEE, GCAM, IMAGE, MESSAGEix-GLOBIOM, and REMIND-MAgPIE. In the mid-2000s, this class of models began evolving to include multi-sector (and multi-model) interactions and feedbacks at variable (and finer) spatiotemporal resolution, enabling analysis of impacts, adaptation, and vulnerability (IAV) under global change. To distinguish this new focus, some of these models (e.g., GCAM) are now also categorized as Multi-Sector Dynamics (MSD) models. The representation of water resources in these models grew rapidly alongside the community’s increasing focus on IAV. Yet there is a lack of clarity around the similarities and differences across the community’s models with regard to water resources, which poses challenges for their application in more detailed disciplinary contexts (e.g., water resources planning and management). Here we ask: Do integrated models (including detailed-process IAM and MSD models with global coverage) share elements of a common philosophy on modeling water resources, and how can it be described? To address this, we compare similarities and differences across a representative set of global integrated models along three dimensions: supply, demand, and supply-demand interactions. We find that similarities across models exist mostly on the supply side, as most explicitly represent global gridded hydrology. However, many differences across models arise in the representation of human water demands and supply-demand interactions, especially regarding water’s interpretation as a physical versus economic commodity, the demand sectors included and their representation, rules used to allocate scarce water across demands, the role of infrastructure, and approaches to validating the models’ human dimensions. When complete, this analysis will enable the articulation of the elements to a common philosophy, and more broadly will provide value to the integrated modeling community as it seeks to grow its water resources research contributions and establish its niche relative to other water research communities.