TLDR: Comparing the speed of adiabatic quantum computers with unitary quantum computers is rather difficult, and even the physics behind the two hint that the optimum approach to increased speed is different for the two cases.

Even between different implementations of unitary quantum computers, speed comparisons are not so clear. Gate speeds can be compared, but that doesn't tell you a lot about how fast you will get a reliable result from a computation. It also doesn't offer much insight into the future.

In short, don't go betting on an architecture just yet.