FME AR models can be stored in public locations (e.g. GoogleDrive or DropBox). They can also be stored locally on the device.

In some scenarios, the data is getting obsolete very rapidly and building a model with an up-to-date data is critical. In this case, a model can be created on request when it is needed - rather than ahead of time. This can be achieved by using FME Data Express to submit a job request (i.e. run a workspace on FME Server/FME Cloud) that creates a model and streams it back to the user's device. The model will be streamed directly into FME AR application.