Easy answer to this; None. All hexbeams are kings.
All hexbeams built after the broadband model invented by G3TXQ perform on air about the same. You can run computer models and get gain figures for free space (which is what most antenna manufacturers do) and it will be close on all hexbeams. But if you do the same computer model for the antenna over ground, the same hexbeam will show a much higher gain. Why is this? Well, in the real world, antennas are above ground and there is a reflection of the radio wave that bounces back up and combines with the main wave to increase its power. This makes the gain higher.
But the amount of this reflection depends on how high above the ground the antenna is and the vertical attack angle of the radio wave, and the type of ground. To make any comparison you have to explain how high the antenna is and what vertical attack angle you are using, yahdah, yahdah. As a result of this complexity, most antenna manufacturers state their gain in terms of the theoretical gain in free space with no ground. The reason for this is that you can then compare antennas more easily. For example, a dipole always has a free space gain of about 2 dB and a two element beam has a free space gain of about 5 dB. That’s it. You can see immediately which is better. It doesn’t matter whether it is over salt water or good ground or rocky ground or how high. With free space comparisons you are comparing just the antennas and not the environment they are in.
The same is true with hexbeams. If you have a gain of 9 dB with one hexbeam and 5 dB with another, which one is better? It depends on what measurement method you use. The “above ground” gain of a hexbeam is about 9-10 dB even though the free space gain is only about 5.5 dB. The actual performance is the same; you are just using two different measuring sticks. If you are going to compare hexbeams, you need to use the same measuring stick. And it turns out that the differences of various hexbeams are tiny and mostly have to do with how well you built the computer model rather than the antennas themselves.
The main differences among hexbeams are in the physical construction details. How robust is the hexbeam? How will it fare in salt water environments, wind, snow and ice? Sturdiness and survivability are important but you can make a hexbeam like an M1 Abrams tank and it will indeed last. But what is the cost and is it too heavy to mount it on an inexpensive mast or do you have to have a broadcast quality tower to support it? As with all good engineering, it is a matter of balancing cost vs survivability.
Look for the best balance of price and structural integrity.