FP8 vs BF16 on H100: what a million tokens costs on AWS, Azure and GCP in 2026
vLLM's April 2026 benchmark put FP8 KV-cache 14.9% above BF16 on one H100. Priced out, that is $4.24 down to $3.69 per million output tokens - and your AWS purchase model moves the number four times further.
