Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) on DeepSeek R1. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics β same interactions as the main inference chart.
Throughput at 50 tok/s/user on DeepSeek R1: B200 hits 4852 tok/s/chip, H200 hits 1591. Per-million costs land at $0.10 and $0.21 respectively. B200 is 115% cheaper per token; B200 delivers 205% more tok/s/chip.
B200 / H200 on DeepSeek R1 at 89 tok/s/user: 1461 / 819 tok/s/chip, $0.33 / $0.41 per million tokens. B200 is 26% cheaper per token; B200 delivers 78% more tok/s/chip.
Toward the upper edge of the 12β167 tok/s/user interactivity band, at 128 tok/s/user on DeepSeek R1: B200 runs 1051 tok/s/chip at $0.46/M tokens, H200 runs 625 at $0.54/M. B200 is 19% cheaper per token; B200 delivers 68% more tok/s/chip. (Numbers reflect the default 8k/1k Β· fp8 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:4851.9H200:1590.9 | B200:1461.3H200:819.4 | B200:1051.0H200:624.5 |
| Cost ($/M tok) | B200:$0.099H200:$0.213 | B200:$0.329H200:$0.414 | B200:$0.457H200:$0.543 |
| tok/s/MW | B200:2837367H200:1161268 | B200:854547H200:598071 | B200:614600H200:455850 |
| Concurrency | B200:~292H200:~73 | B200:~34H200:~14 | B200:~8H200:~4 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr:
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 β’ Drag to pan β’ Double-click to reset β’ Click a point to pin tooltip
DeepSeek R1 0528 671B 8K / 1K Total Tokens per $1 TCO vs. Interactivity