There isn’t energy cost between bodies, but there are huge energy costs getting into orbit, leaving orbit, and getting into the destination body’s orbit
> getting into the destination body's orbit
I said round-trip energy costs. The delta-v for arriving cargo to enter orbit equals the delta-v for departing cargo. Over a full cycle of sending equal payloads both ways, the net change of energy and momentum is zero. I linked to an analysis of one way to harness this. When the skyhook in the linked example boosts a payload from Venus-Earth Hohmann to Earth-Mars Hohmann, the skyhook's orbit gets perturbed; but it gets the opposite perturbation on the boost on the opposite traffic.
(This is not the only non-rocket approach to free-space transport with potentially low marginal cost.)
Niche applications like low-temperature ZBLAN manufacturing might make more sense.