Comment by finnh
The energy densities listed are flagged as approximate, so grains of salt etc, but the numbers on the page aren't entirely consistent.
The stated energy density is "> 500 watthours/liter".
But higher on the page we see a relative-energy-density bar graph shows lightcell at 5x the energy density of lithium batteries, and (38/5 =) 7.6x less dense then petrol. This implies an energy density for lightcell of 1250 Wh/liter, as (according to Google) petrol clocks in just under 9500 Wh/liter, and (again according to Google) lithium batteries can reach 300 Wh/liter so let's call it 250 for the math to work out.
I'm curious which number is closer to truth: 500Wh/liter, or 1250? Is 1250 the theoretical max and 500 the current output in a test rig?
I believe the bar graph is showing relative energy densities of the raw energy sources so the 5x bar is just the energy density of hydrogen as H2. Your 1250 Wh/L number is right for compressed gaseous hydrogen so The 500Wh/L lines up with burning H2 at 40% efficiency. The "use fuel for extended duration" implies that they believe they can achieve a much higher Wh/L with other fuels.