MiniMax M3 428B — B200 vs B300 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus B300 (NVIDIA Blackwell) on MiniMax M3 428B. 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 121 tok/s/user on MiniMax M3 428B, B200 costs $0.08 per million tokens; B300 costs $0.11. B200 is 28% more cost-efficient at this operating point.
B200 edges B300 at 235 tok/s/user on MiniMax M3 428B — $0.15 per million tokens versus $0.20, a 33% cost-per-token gap.
Push MiniMax M3 428B to 349 tok/s/user and B200 lands at $0.29 per million tokens against B300's $0.31 — B200 pulls ahead by 6%. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Chip pricing (owning hyperscaler): B200 $1.73/chip/hr · B300 $2.26/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B200:$0.084B300:$0.107 | B200:$0.151B300:$0.201 | B200:$0.293B300:$0.311 |
| Concurrency | B200:~20B300:~13 | B200:~21B300:~8 | B200:~12B300:~3 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.