I was leaning towards going 1ml per vial until I read Bill Roberts article and the information regarding how much BA to use for reconstituting:
A problem with using small amounts of water can be that after drawing everything possible from a vial, a substantial percentage of it may remain, causing significant loss of GH. The loss can be about 10%.
One solution is this: Reconstitute your next vial by first adding the needed amount of water to the near-empty vial, then drawing it, and finally adding to the new vial. This will transfer nearly all the remaining GH to the new vial. There is no need to worry about resulting dosing error: over time everything averages out.
If that method does not appeal, another solution is to use a larger amount of water for reconstitution. For example, 2.4 mL of water can be used for an 8 IU vial. Each 0.3 mL of injection then provides 1 IU. And if the last 0.1 mL cannot be drawn from the vial, this is them only about a 4% loss. That’s still a waste, but it’s much reduced compared to reconstituting with less water.