Without the Space Shuttle in 2018, the ISS could not be built as it currently exists.
The Russians would likely just build their segment the same way. Launching major modules on Proton (Zarya, Zvezda). The minor modules launched on a Soyuz booster using a modified Progress module for controlling the flight.
You can tell this is the case, since the Russian plan for their next station is to use the next two ISS Modules (Nuaka and OM) in 2024 (or whenever) to form the core of their next station, with additional modules launched as needed.
US side a Falcon Heavy could launch a lot of mass, and SpaceX could probably derive the Dragon propulsion system onto a satellite bus style approach, to deliver the modules.
But there will be no construction base with a manipulator arm, that was there with the Shuttle and CanadaArm.
Thus the design would be quite different in all likelyhood, and in fact almost with certainty.
Launch costs by Falcon Heavy would be hugely cheaper than by Space Shuttle, which could help a lot of the costs.
Bigelow would likely be able to provide more modules, and when building several, reduce costs further. But more so, they would charge for a module, and deliver it.
If you waited a couple of years from now, when the BFR/BFS is in flight launch costs would be even cheaper.
But how much did flight costs dominate ISS costs?
Absolutely it has a large affect, but I suspect that if you could ditch the ISS management overhead and run it simpler (I.e. Get rid of NASA entirely in running it) it would be much cheaper and a much larger component than the launch costs.
Conversely if you look at cost of the Shuttle, not very much of it (of the whole) was actually materials costs. Rather it was the 24,000 someodd employees of NASA working on the programs salaries that were fixed cost each year.
I.e. The key is that the cost of the ISS is less about materials and almost entirely about salaries. Ditch the thousands (tens of thousands?) employees of NASA and replace it with a leaner company to run it, and overall cost will be much cheaper.
Thus short answer - no way to compare. Longer answer, if done smart and not the NASA big government way, much cheaper.