DeepSeek V4 Pro 1.6T — GB300 NVL72 vs H200 Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus H200 (NVIDIA Hopper) 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.
GB300 NVL72 costs $0.08 per million tokens at 72 tok/s/user on DeepSeek V4 Pro 1.6T; we have no H200 benchmark data at this exact target.
At 130 tok/s/user on DeepSeek V4 Pro 1.6T, GB300 NVL72 comes in at $0.22 per million tokens. H200 hasn't been benchmarked at this operating point.
Only GB300 NVL72 has cost data at 187 tok/s/user on DeepSeek V4 Pro 1.6T — $1.44 per million tokens. H200 is unmeasured at this 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.)
GPU pricing (owning hyperscaler): GB300 NVL72 $2.65/GPU/hr · H200 $1.41/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 | GB300 NVL72:$0.078H200:— | GB300 NVL72:$0.216H200:— | GB300 NVL72:$1.436H200:— |
| Concurrency | GB300 NVL72:~1163H200:— | GB300 NVL72:~288H200:— | GB300 NVL72:~13H200:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.