MiniMax M2.5/M2.7 — B200 vs B300 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus B300 (NVIDIA Blackwell) 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.
B200 edges B300 at 69 tok/s/user on MiniMax M2.5/M2.7 — $0.07 per million tokens versus $0.10, a 42% cost-per-token gap.
Push MiniMax M2.5/M2.7 to 112 tok/s/user and B200 lands at $0.22 per million tokens against B300's $0.26 — B200 pulls ahead by 19%.
B200: $0.72 per million tokens. B300: $0.83. Both at 155 tok/s/user on MiniMax M2.5/M2.7, with B200 16% cheaper. (Numbers reflect the default 1k/1k · fp4 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 · B300 $2.34/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:$0.068B300:$0.097 | B200:$0.222B300:$0.264 | B200:$0.719B300:$0.834 |
| Concurrency | B200:~711B300:~394 | B200:~128B300:~95 | B200:~11B300:~13 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.