DeepSeek V4 Pro 1.6T — GB300 NVL72 vs MI325X Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus MI325X (AMD CDNA 3) on DeepSeek V4 Pro 1.6T. 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 71 tok/s/user on DeepSeek V4 Pro 1.6T, GB300 NVL72 comes in at $0.07 per million tokens. MI325X hasn't been benchmarked at this operating point.
Only GB300 NVL72 has cost data at 129 tok/s/user on DeepSeek V4 Pro 1.6T — $0.18 per million tokens. MI325X is unmeasured at this target.
GB300 NVL72 costs $1.28 per million tokens at 187 tok/s/user on DeepSeek V4 Pro 1.6T; we have no MI325X benchmark data at this exact target. (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): GB300 NVL72 $2.31/chip/hr · MI325X $1.10/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:$0.067MI325X:— | GB300 NVL72:$0.183MI325X:— | GB300 NVL72:$1.279MI325X:— |
| Concurrency | GB300 NVL72:~1113MI325X:— | GB300 NVL72:~291MI325X:— | GB300 NVL72:~13MI325X:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.