So a Tesla using a heat pump means it consumes much less energy when heating up a cabin than resistors and, pertinently to my case, will consume an amount of electricity similar to a normal A/C (defined by narf as about 1kw or so) when cooling it, yes?
My old tesla still has the resistive heater, which will pull up to 7 kW (I think?) of electrical power
Usually, aside from crazy outside temperatures, you can think of current heat pumps as working with a coefficient of performance (COP, because "efficiency" > 100% sounds wrong) of about 3-ish, meaning for those 7 kW of heat inside, it'll consume about 2.3 kW electrical power [edit: to clarify, this is only for a spike, like with the A/C - so I'd expect the heat pump to consume on an average cold day way less than 1 kW average over a longer drive, i think there's a teslabjörn video about this
somewhere, if you're interested]. When it gets too cold (like -10C or below I'd say) this COP usually plummets (depending on system design), hence there is a resistive heater in front of the heat pump sometimes.
The A/C should operate as narf described, so high electrical draw when first cooling down the car, then settle down somewhere at maybe 1 kW average over a longer drive.
In addition to the heatpump however, the new tesla design has the crazy OCTOVALVE stuff - basically they have a large heat exchanger in a central position where EVERYTHING that's warm or gets cooled or anything can be made to exchange heat with each other. So in addition to having a more efficient heater, they actually harvest heat (or dump heat when they need to get rid of it I guess) from all available components (i.e. motors, battery, whatever). This further improves things, but I have no idea and no way of telling by how much