[comicjk]

These neural network models do not need to know whether a bond exists - in fact, they have no concept of bond topology. They are designed to be a drop-in replacement for DFT in terms of energies and forces. The only inputs are XYZ coordinates and chemical element labels for the nuclei (and, in the near future, net charge of the system).

[fock]

with the non-bonded interactions parametrized with dimers it might work... sometimes (might be good enough for a lot of things though)