Seasonal snow responds to climate variability and long-term changes, as it is influenced by temperature and precipitation patterns. Characterizing this variability is done using a combination of historical observations and model experiments. Coupled ("online") and uncoupled ("offline") model experiments can be testbeds to investigate the seasonal snowpack’s response to perturbations, but the model spread needs to be understood to address uncertainties, especially for planning and impacts mitigation. I will show results from a study of Northern Hemisphere snow which uses snow outputs from 12 CMIP6 climate models. We use the offline Brown Temperature Index Model (B-TIM), driven by temperature and precipitation data from the same 12 models, to produce a set of snow reconstructions which can be compared to each other and to the raw CMIP6 snow output. We assess mean and peak snow water equivalent (SWE), snow cover, trends, and spatial patterns and relate them to temperature and precipitation biases. These insights support benchmarking efforts and provide new insight on snow change projections to the end-of-century (under the SSP5-8.5 scenario).