MiniMax M3 428B — GB300 NVL72 vs H100 Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus H100 (NVIDIA Hopper) 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.
GB300 NVL72: $0.04 per million tokens. H100: $0.33. Both at 75 tok/s/user on MiniMax M3 428B, with GB300 NVL72 680% cheaper.
Around the middle of the 22–235 tok/s/user interactivity band — at 129 tok/s/user — GB300 NVL72 runs $0.16 per million tokens on MiniMax M3 428B while H100 runs $0.46. GB300 NVL72 is the cheaper choice by 181%.
On MiniMax M3 428B at 182 tok/s/user, the per-million math comes out to $0.29 for GB300 NVL72 and $0.60 for H100; GB300 NVL72 delivers 104% more output per dollar. (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.)
Chip pricing (owning hyperscaler): GB300 NVL72 $2.31/chip/hr · H100 $1.17/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 | GB300 NVL72:$0.042H100:$0.328 | GB300 NVL72:$0.164H100:$0.462 | GB300 NVL72:$0.295H100:$0.602 |
| Concurrency | GB300 NVL72:~2048H100:~13 | GB300 NVL72:~55H100:~6 | GB300 NVL72:~11H100:~3 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.