At the turn of the century R. G. Roble advanced an ambitious program of developing an atmospheric general circulation model (GCM) extending from the surface to the exosphere. He outlined several areas of research and application to potentially benefit from what is now commonly called whole atmosphere modeling. The purpose of this article is to introduce this new field to a broader geophysical community and document its progress over the last decade. Vertically extended models are commonly built from existing weather and climate GCM codes incorporating a number of approximations, which may no longer be valid. Promising directions of further model development, potential applications, and challenges are outlined. One application is space weather or day-to-day and seasonal variability in the ionosphere and thermosphere driven by meteorological processes from below. Various modes of connection between the lower and upper atmosphere had been known before, but new and sometimes unexpected observational evidence has emerged over the last decade. Persistent “nonmigrating” wavy structures in plasma and neutral densities and a dramatic response of the equatorial ionosphere to sudden warmings in the polar winter stratosphere are just two examples. Because large-scale meteorological processes are predictable several days in advance, whole atmosphere weather prediction models open an opportunity for developing a real forecast capability for space weather.