the API isn’t really designed to get more than a few tens of thousands of records. so if by “whole global dataset” you mean all observations with observation fields, i don’t think there’s a good way to get that much data.
if you’re looking for interactions associated with specific taxa, that might make it more feasible (if there are few enough observations) to get that data via API or other means.
some data from iNaturalist does end up in globalbioticinteractions.org, and you can download the data they have over there. that might be your best bet if you’re trying to analyze interactions.