Comment by 332451b

I assume this is idealized somehow, given how close it is to DeepInfer numbers at 100% utilization. My point was more that there is seemingly a lot of margin to absorb underutilization, outages and other costs.