We analyze the ability of the NARCCAP ensemble of regional climate models (RCMs) to simulate extreme monthly precipitation and its supporting circulation for regions of North America, comparing 18 years of simulations driven by the NCEP-DOE reanalysis with observations. Analysis focuses the wettest 10% of months during the cold half of the year (October-March), when we assume that synoptic circulation governs precipitation. For a coastal California region where the precipitation is largely topographic, the models individually and collectively replicate well the months of greatest extremes, the amount of extreme precipitation and the 500 hPa circulation anomaly associated with the extremes. The simulated top 10% wettest months average less than 3% more precipitation than the observed wettest 10%. The models also replicate very well the statistics of the interannual variability of occurrences of extremes: 98% of all simulated monthly extremes among the models occur in a year with an observed extreme. For an interior region containing the Upper Mississippi River Basin, where precipitation is more dependent on internally generated storms, the models still show agreement with observations, though not as closely. Circulation anomalies for extreme months are roughly similar in the models and observations, and the simulated top 10% wettest months average about 5% less precipitation than observed wettest 10%. The models also replicate fairly well for this region the interannual variability of extreme monthly precipitation: 80% of all simulated monthly extremes among the models occur in a year with an observed extreme.