The predictions of hadronic interaction models for cosmic-ray induced air showers contain inherent uncertainties due limits of the accelerator data and theoretical understanding in the required energy and rapidity regime. Model uncertainties are typically evaluated in the range appropriate for cosmic-ray air shower arrays (10$^{15}$-10$^{20}$ eV), however the performance of the models for gamma-ray observatories is becoming more and more important.
We assess the model differences on the gross behaviour of the predictions in the energy (0.1-100 TeV) and altitude ranges most appropriate for detection by current ground-based gamma-ray observatories. We go on to investigate the particle production at the first interaction point to extrapolate how the differences in micro-physics of the models may compound into differences in the gross air shower behaviour.