Hi @Francisco,
That’s too bad that they don’t tell you the primer set as that bit of information is pretty useful in a few steps of a regular pipeline. The safest bet would be to use the full 16S classifier that are available from the data resource page. These will be slightly less accurate than specific-region classifiers but not too much that would be a bother. You can also always just use a typical V3-V4 classifier (there’s some pre-trained ones for Greengenes and Silva floating on the community contributions section of the forum. You’ll know pretty quickly if the region is off, you would see some poor classifications. But yeah, safe set, full 16S.