I see it as they did everything they can to compare the specific code path. If your workload scales with FP16 but not with tensor cores, then this is the correct way to test. What do you need for LLM inference?
Couldn't they find a real workload that does this?
vLLM inference of Mixtral in fp16 is a real workload. I guess the details are there because of the different inference engine used. You need the most similar compute tasks to be ran but the compute kernels can't be the same as in the end they need to be ran by a different hardware.