Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) 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 / H200 on gpt-oss 120B at 86 tok/s/user: 37093 / 6958 tok/s/chip, $0.01 / $0.05 per million tokens. B200 is 276% cheaper per token; B200 delivers 433% more tok/s/chip.
Around the middle of the 30–255 tok/s/user interactivity band, at 142 tok/s/user on gpt-oss 120B: B200 runs 25056 tok/s/chip at $0.02/M tokens, H200 runs 4653 at $0.07/M. B200 is 280% cheaper per token; B200 delivers 438% more tok/s/chip.
Setting 199 tok/s/user as the target on gpt-oss 120B, B200 produces 17264 tok/s/chip ($0.03 per million tokens) and H200 produces 2879 ($0.12). B200 is 323% cheaper per token; B200 delivers 500% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:37092.9H200:6957.7 | B200:25055.8H200:4653.1 | B200:17264.5H200:2878.6 |
| Cost ($/M tok) | B200:$0.013H200:$0.049 | B200:$0.019H200:$0.073 | B200:$0.028H200:$0.118 |
| tok/s/MW | B200:21691776H200:5078638 | B200:14652486H200:3396459 | B200:10096180H200:2101202 |
| Concurrency | B200:~46H200:~25 | B200:~32H200:~16 | B200:~16H200:~7 |
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