MiniMax M3 428B — GB200 NVL72 vs MI355X Performance per Dollar
Cost per million tokens of GB200 NVL72 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) 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.
Push MiniMax M3 428B to 66 tok/s/user and GB200 NVL72 lands at $0.13 per million tokens against MI355X's $0.08 — MI355X pulls ahead by 51%.
GB200 NVL72: $0.27 per million tokens. MI355X: $0.12. Both at 106 tok/s/user on MiniMax M3 428B, with MI355X 132% cheaper.
Toward the upper edge of the 27–186 tok/s/user interactivity band — at 147 tok/s/user — GB200 NVL72 runs $0.64 per million tokens on MiniMax M3 428B while MI355X runs $0.17. MI355X is the cheaper choice by 279%. (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): GB200 NVL72 $1.86/chip/hr · MI355X $1.50/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 | GB200 NVL72:$0.128MI355X:$0.084 | GB200 NVL72:$0.270MI355X:$0.117 | GB200 NVL72:$0.637MI355X:$0.168 |
| Concurrency | GB200 NVL72:~222MI355X:~40 | GB200 NVL72:~45MI355X:~18 | GB200 NVL72:~13MI355X:~9 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.