MiniMax M2.5/M2.7 — B200 vs H100 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus H100 (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.
On MiniMax M2.5/M2.7 at 44 tok/s/user, the per-million math comes out to $0.06 for B200 and $0.19 for H100; B200 delivers 238% more output per dollar.
At 66 tok/s/user on MiniMax M2.5/M2.7, B200 costs $0.11 per million tokens; H100 costs $0.29. B200 is 155% more cost-efficient at this operating point.
B200 edges H100 at 89 tok/s/user on MiniMax M2.5/M2.7 — $0.17 per million tokens versus $0.44, a 158% cost-per-token gap. (Numbers reflect the default 8k/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): B200 $1.95/GPU/hr · H100 $1.30/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 | B200:$0.057H100:$0.192 | B200:$0.114H100:$0.291 | B200:$0.170H100:$0.439 |
| Concurrency | B200:~512H100:~19 | B200:~21H100:~16 | B200:~20H100:~4 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.