Noting today, from the Jan 10th, 00Z FV3 at the 90-hr forecast time step, the FV3 has more snow extending south across central TN and even into northern Alabama.  Some of the difference appears to be a timing issue, as in the images further below, from model runs 24- and 48- hours earlier, the GFS has higher amounts further south, which it no longer has.

I also have seen that, in general, the FV3 seems to be further east than the GFS with its axis of heaviest snow--more snow shown for western PA, than the GFS, which can be seen by viewing DModel/Dt views.

Lastly, note in this Jan 10, 00Z run, that the FV3 has an area of >30" in 24hrs inland and east of Buffalo (not lake effect) vs 12-14" max from the GFS for this 24-hr time step ending 18Z, Jan 13.  30" in 24 hours quite hefty, especially for non-Lake Effect!!  

Indeed the run 24 hours earlier, and also 48 hours earlier, both show the GFS with snow further south than the FV3, as shown here and here.

Now, 24 hours later, using the 66-hr forecast from the 00Z, Jan 11th run, it has backed off on the higher snow amounts in TN and into AL.  Now none in AL, and just a tad in central TN.