“Are you sure siphons don’t work in a vacuum?”
I question this too. The atmospheric pressure is exactly equal at both ends of the tube. The only reason it might not work in a “perfect” vacuum is that many liquids boil in vacuum.
Think about it; the receiving end is lower, so it is under greater ambient pressure than the higher, sending end.
Assuming that both the ambient pressure and experienced gravity are nonzero, the gradient in ambient pressure works to impede the flow, but only very slightly compared to the flow-creating differential due to the weight difference in the two halves of the siphon.
The difference in the magnitude of the two effects is proportional to (1) the strength of the gravity, and (2) the difference in mass density between the ambient air (or other gas) and the liquid, and (3) the net length of the drop.