A great need exists to develop biomass pyrolysis models to predict pyrolysis products and yields as a function of different feedstocks and reaction conditions. As a first step, an attempt has been made to rapidly characterize different types of biomass using the micro-pellet reactor, laser pyrolysis-molecular beam mass spectrometer (MBMS) and factor analysis. In this work, we used the techniques that were developed in our laboratory to show the capability of rapid characterization of various biomass feedstocks. All pellets were pyrolyzed with lower laser power to minimize secondary reactions. A molecular beam mass spectrometer (MBMS) allows direct and real time sampling from a pyrolysis system. Multivariate factor analysis was employed to simplify the data and discover underlying differences in chemistry that distinguish one type of biomass from others due to its characteristics.