It's the initial bolus of hydrogen to re-wet the Venus atmosphere and cool it off that's the megaproject. After that, the replacement rate is ridiculously low.

It's the initial bolus of hydrogen to re-wet the Venus atmosphere and cool it off that's the megaproject. After that, the replacement rate is ridiculously low. And if the lost hydrogen is never replaced, the terraforming-added hydrogen will still likely last longer than human civilization. Venus has been losing H2 and He for billions of years, and still has some left to lose. If we add enough H2 to precipitate all the CO2 out as mountains of C and oceans of H2O, and cool off the planet enough that the hydrogen stays in the water, the atmosphere will be fine for the foreseeable future. One basic Venus atmospheric-terraforming equation is "2 CO2 + MgH2 -> C + MgCO3 + H2O", by multiple reactions at different temperatures and pressures. By mass, 4.6e23 g CO2(g) + 1.4e23 g MgH2(s) -> 6.3e22 C(s) + 4.4e23 g MgCO3(s) + 9.4e22 g H2O(l). That produces an ocean about 12% the mass of Earth's, but requires roughly 4% the mass of the asteroid belt in MgH2, or a sphere of solid MgH2 with diameter 570 km--about the same volume as Pallas, Vesta, or Hygeia (but half the mass). Magnesium is the 12th-most abundant element in the solar system, and could be mined from olivine-plagioclase asteroids. (Calcium can also be used to transport H2.) About 17% of asteroids are S-type. Seems doable. Alternatively, decompose the MgH2 to Mg and H2 in Venus orbit, and only send down the H2, sending the Mg back out to store more hydrogen. That's "CO2 + 2 H2 -> C + 2 H2O". By mass, that's 4.6e23 g CO2(g) + 4.2e22 g H2(g) -> 1.3e23 C(s) + 3.8e23 g H2O(l), which would make an ocean about 28% as massive as Earth's, but requires that some magnesium (or calcium) make several round trips just to carry H2, which is relatively light. If you're going to do that, you could also just ship H2O to Venus, electrolyze it in orbit with solar power, drop the H2 down the gravity well, and ship the O2 back out to the comet wranglers for breathing gas. You'd need to move exactly as much water into orbit as you're eventually creating on the surface, so 3.8e23 g H2O. That's more mass to ship than the MgH2 answer, but at least we know that the water is easily found in comets, and on Europa and Callisto, and easily processed. There's no reason why any of these should be mutually exclusive.