gpt-oss 120B — GB200 NVL72 vs H100 Performance per Dollar
Cost per million tokens of GB200 NVL72 (NVIDIA Blackwell) versus H100 (NVIDIA Hopper) on gpt-oss 120B. 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.
GB200 NVL72 edges H100 at 117 tok/s/user on gpt-oss 120B — $0.03 per million tokens versus $0.12, a 353% cost-per-token gap.
Push gpt-oss 120B to 166 tok/s/user and GB200 NVL72 lands at $0.03 per million tokens against H100's $0.24 — GB200 NVL72 pulls ahead by 650%.
GB200 NVL72: $0.06 per million tokens. H100: $0.44. Both at 216 tok/s/user on gpt-oss 120B, with GB200 NVL72 626% cheaper. (Numbers reflect the default 1k/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): GB200 NVL72 $1.86/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 | GB200 NVL72:$0.027H100:$0.124 | GB200 NVL72:$0.031H100:$0.236 | GB200 NVL72:$0.061H100:$0.439 |
| Concurrency | GB200 NVL72:~264H100:~64 | GB200 NVL72:~2684H100:~17 | GB200 NVL72:~102H100:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.