A $50 billion Nvidia inference factory will generate lower cost tokens than a $30-40 billion alternative due to 10x throughput difference

It is very likely that the $50 billion factory, and in fact, I can prove it that the $50 billion factory will generate for you the lowest cost tokens.

That is not a large percentage when the $50 billion data center is 10 times the throughput,