All probabilistic genotyping (PG) software used to assist with the interpretation of forensic DNA profiles make modelling assumptions to help interpret forensic DNA profiles. However, not all PG software are the same, different PG software may make different assumptions or implement different models. Moreover, it is known that different models implemented in different software can produce divergent results. This thinking is reflected in the 2016 PCAST (President’s Council of Advisors on Science and Technology) report, where the authors called for (amongst other things) an investigation into “Under what circumstances – and why – does the method produce results (random inclusion probabilities) that differ substantially from those produced by other methods?”. However, with a good understanding of each software, the differences arising from these assumptions can be predicted; and in some cases, software options or workarounds allow these differences to be minimised. This presentation aims to summarise the results from a comparison of LRs obtained from the interpretation of more than 120 PROVEDIt profiles using two probabilistic genotyping software – EuroForMix and STRmix™. The results from both software show similar LR behaviour, suggesting that even though the models in the two software are different, both can be useful in assigning an LR. Any difference in the assigned LR between the software was investigated to try and identify the driving factor(s).