COSMO-RS (conductor-like screening model for real solvents) and three different group-contribution methods were used to compute saturation (subcooled) liquid vapor pressures for 16 possible products of ozone-initiated α-pinene autoxidation, with elemental compositions C10H16O4-10 and C20H30O10-12. The saturation vapor pressures predicted by the different methods varied widely. COSMO-RS predicted relatively high saturation vapor pressures values in the range of 10(-6) to 10(-10) bar for the C10H16O4-10 "monomers", and 10(-11) to 10(-16) bar for the C20H30O10-12 "dimers". The group-contribution methods predicted significantly (up to 8 order of magnitude) lower saturation vapor pressures for most of the more highly oxidized monomers. For the dimers, the COSMO-RS predictions were within the (wide) range spanned by the three group-contribution methods. The main reason for the discrepancies between the methods is likely that the group-contribution methods do not contain the necessary parameters to accurately treat autoxidation products containing multiple hydroperoxide, peroxy acid or peroxide functional groups, which form intramolecular hydrogen bonds with each other. While the COSMO-RS saturation vapor pressures for these systems may be overestimated, the results strongly indicate that despite their high O:C ratios, the volatilities of the autoxidation products of α-pinene (and possibly other atmospherically relevant alkenes) are not necessarily extremely low. In other words, while autoxidation products are able to adsorb onto aerosol particles, their evaporation back into the gas phase cannot be assumed to be negligible, especially from the smallest nanometer-scale particles. Their observed effective contribution to aerosol particle growth may therefore involve rapid heterogeneous reactions (reactive uptake) rather than effectively irreversible physical absorption.