MiniMax M3 428B — B200 vs GB300 NVL72 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus GB300 NVL72 (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.
Only B200 has cost data at 148 tok/s/user on MiniMax M3 428B — $0.09 per million tokens. GB300 NVL72 is unmeasured at this target.
B200 costs $0.22 per million tokens at 290 tok/s/user on MiniMax M3 428B; we have no GB300 NVL72 benchmark data at this exact target.
At 432 tok/s/user on MiniMax M3 428B, B200 comes in at $0.58 per million tokens. GB300 NVL72 hasn't been benchmarked at this operating point. (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 · GB300 NVL72 $2.31/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:$0.087GB300 NVL72:— | B200:$0.217GB300 NVL72:— | B200:$0.578GB300 NVL72:— |
| Concurrency | B200:~40GB300 NVL72:— | B200:~4GB300 NVL72:— | B200:~2GB300 NVL72:— |
Cost per Million Total Tokens (Owning - Hyperscaler) vs. Interactivity
MiniMax M3 428B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
MiniMax M3 428B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™TCO $/chip/hr: VR200: 3.61H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68Jalapeño (Teacup): 1.47
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
No data available
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
Interactivity is the rate at which a single user receives generated tokens while the model streams its answer — how quickly new words appear on screen. Higher values feel snappier; operators trade it against batch throughput.
Estimated infrastructure cost of producing one million total tokens (input + output) at this operating point, priced with the all-in hourly ownership cost of a hyperscaler operator. Lower is cheaper.
Formula: $/Mtok = all-in cost per chip-hour ($) × 1,000,000 ÷ (3,600 × total tok/s/chip)