ORE adds new QuantLib-style models/processes/engines, market evolution methods, collateral modelling for variation margin and dynamic initial margin etc in three libraries (QuantExt, OREData, OREAnalytics). And it acts as a configurable application around these three libraries including QuantLib.
Please have a look at the FAQs and the user guide (see the link on the documentation page). That should give you an overview over the project and the relation to QuantLib.
this way we are independent of the QuantLib release cycle and can introduce missing pieces whenever we need them. Another reason is that some classes in QuantExt may seem too specialised for QuantLib, like the InterpolatedDiscountCurve variant which was optimised for the XVA simulation run.
But it’s not impossible that the two libraries might be merged in the future, let’s see.