MiniMax M2.5/M2.7 — GB200 NVL72 vs H200 Performance per Dollar
Cost per million tokens of GB200 NVL72 (NVIDIA Blackwell) versus H200 (NVIDIA Hopper) on MiniMax M2.5/M2.7. Owning-hyperscaler TCO normalized by output tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
At 52 tok/s/user on MiniMax M2.5/M2.7, GB200 NVL72 costs $0.17 per million tokens; H200 costs $0.34. GB200 NVL72 is 99% more cost-efficient at this operating point.
GB200 NVL72 edges H200 at 80 tok/s/user on MiniMax M2.5/M2.7 — $0.50 per million tokens versus $0.68, a 37% cost-per-token gap.
Push MiniMax M2.5/M2.7 to 107 tok/s/user and GB200 NVL72 lands at $0.92 per million tokens against H200's $1.27 — GB200 NVL72 pulls ahead by 37%. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
GPU pricing (owning hyperscaler): GB200 NVL72 $2.21/GPU/hr · H200 $1.41/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | GB200 NVL72:$0.169H200:$0.337 | GB200 NVL72:$0.497H200:$0.683 | GB200 NVL72:$0.925H200:$1.271 |
| Concurrency | GB200 NVL72:~489H200:~106 | GB200 NVL72:~64H200:~14 | GB200 NVL72:~32H200:~5 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.