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 59 tok/s/user on MiniMax M3 428B, B200 costs $0.63 per million tokens; B300 costs $0.39. B300 is 62% more cost-efficient at this operating point.
B300 edges B200 at 106 tok/s/user on MiniMax M3 428B — $1.42 per million tokens versus $1.61, a 14% cost-per-token gap.
Push MiniMax M3 428B to 152 tok/s/user and B200 lands at $4.32 per million tokens against B300's $2.83 — B300 pulls ahead by 52%. (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): 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:$0.633B300:$0.392 | B200:$1.614B300:$1.421 | B200:$4.317B300:$2.833 |
| Concurrency | B200:~32B300:~63 | B200:~8B300:~20 | B200:~4B300:~7 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.