[ilove196884]

Also TPU v1,v2 and v3 were ASICs, but since v4 they have added some new features so they have a lower performance/watt which is quite near Nvidia's power draw. I think Hopper is at 700W and TPU are around 600W.

[YetAnotherNick]

Power draw doesn't matter in cloud. All that matters is performance/price for the task in hand.

[jeffbee]

In the cloud, opex == energy.

[talldayo]

Which is why you don't send 4 TPUs to do 1 GPU's job.

[jeffbee]

If the TPUs cost less, that implies that they draw less power. If they cost more then nobody will use them.