We will enable pre-fetching per this protocol:

• If we know for certain we’ll need another batch (either we know we will need all documents, or if the previous batch didn't fill up to the requested batchSize), we’ll immediately pre-fetch the next batch, after applying the exponential growth rate as applicable.
• If the maxBound of docs needed is unknown and we’ve already requested three batches, we’ll start pre-fetching upon receipt of the third batch and for all future batches, after applying the exponential growth rate as applicable.