Runoff generation processes in permafrost catchments differ from those in non-permafrost catchments due to a nearly impermeable permafrost boundary and seasonal freeze-thaw of the shallow subsurface, which together modify dominant flowpaths. Permafrost regions, which represent 11% of Earth’s land surface, remain vastly understudied and are rapidly changing due to warming and hydrological intensification, making it critical to understand runoff generation processes in these systems to model and predict future hydrologic and ecological conditions. Here, we synthesize literature on runoff processes in continuous permafrost systems across the Arctic, Antarctic, and Tibetan Plateau to assess which runoff process paradigms are applicable in permafrost catchments and to identify opportunities to advance runoff observations and modeling. We identify when, where, and at what spatial scale runoff processes have been studied to highlight gaps and biases in our understanding, finding that most of the permafrost runoff research is based in North America. We present a conceptual framework for runoff generation organized around four key thaw season periods characterized by distinct subsurface thaw states and energy receipt– snowmelt, early active layer thaw, peak active layer thaw, and freeze-up, the last of which remains largely understudied. Overall, this synthesis demonstrates that studies of runoff generation in continuous permafrost systems have been concentrated at a limited number of sites, missing important variability in precipitation, thaw depth, and soil characteristics across permafrost regions and potentially biasing conceptual understandings of runoff. Finally, we identify key priorities for future observation and modeling efforts across the extensive and rapidly changing permafrost landscapes.