Qwen 3.5 397B-A17B — GB300 NVL72 vs H100 Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus H100 (NVIDIA Hopper) on Qwen 3.5 397B-A17B. 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 H100 has cost data at 63 tok/s/user on Qwen 3.5 397B-A17B — $0.73 per million tokens. GB300 NVL72 is unmeasured at this target.
H100 costs $1.66 per million tokens at 97 tok/s/user on Qwen 3.5 397B-A17B; we have no GB300 NVL72 benchmark data at this exact target.
At 131 tok/s/user on Qwen 3.5 397B-A17B, H100 comes in at $4.06 per million tokens. GB300 NVL72 hasn't been benchmarked at this operating point. (Numbers reflect the default 1k/1k · fp8 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 · H100 $1.30/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:—H100:$0.726 | GB300 NVL72:—H100:$1.664 | GB300 NVL72:—H100:$4.057 |
| Concurrency | GB300 NVL72:—H100:~39 | GB300 NVL72:—H100:~10 | GB300 NVL72:—H100:~3 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.