Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on gpt-oss 120B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
B200 posts 22825 tok/s/chip for $0.02 per million tokens at 155 tok/s/user on gpt-oss 120B; GB200 NVL72 posts 33534 tok/s/chip for $0.02. GB200 NVL72 is 37% cheaper per token; GB200 NVL72 delivers 47% more tok/s/chip.
Throughput at 243 tok/s/user on gpt-oss 120B: B200 hits 13355 tok/s/chip, GB200 NVL72 hits 19526. Per-million costs land at $0.04 and $0.03 respectively. GB200 NVL72 is 36% cheaper per token; GB200 NVL72 delivers 46% more tok/s/chip.
B200 / GB200 NVL72 on gpt-oss 120B at 330 tok/s/user: 6328 / 8905 tok/s/chip, $0.08 / $0.06 per million tokens. GB200 NVL72 is 31% cheaper per token; GB200 NVL72 delivers 41% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:22825.2GB200 NVL72:33534.3 | B200:13355.3GB200 NVL72:19525.7 | B200:6327.9GB200 NVL72:8904.9 |
| Cost ($/M tok) | B200:$0.021GB200 NVL72:$0.015 | B200:$0.036GB200 NVL72:$0.026 | B200:$0.076GB200 NVL72:$0.058 |
| tok/s/MW | B200:13348051GB200 NVL72:17932795 | B200:7810104GB200 NVL72:10441533 | B200:3700528GB200 NVL72:4761975 |
| Concurrency | B200:~32GB200 NVL72:~343 | B200:~6GB200 NVL72:~56 | B200:~8GB200 NVL72:~17 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr: VR200: 3.61H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68Jalapeño (Teacup): 1.47
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
Matching measurements exist, but their chip series are hidden.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
gpt-oss 120B 8K / 1K Total Tokens per $1 TCO vs. Interactivity