Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) 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 — same interactions as the main inference chart.
H100 / H200 on DeepSeek R1 at 45 tok/s/user: 739 / 1773 tok/s/chip, $0.44 / $0.19 per million tokens. H200 is 130% cheaper per token; H200 delivers 140% more tok/s/chip.
Around the middle of the 20–122 tok/s/user interactivity band, at 71 tok/s/user on DeepSeek R1: H100 runs 281 tok/s/chip at $1.16/M tokens, H200 runs 907 at $0.37/M. H200 is 210% cheaper per token; H200 delivers 223% more tok/s/chip.
Setting 97 tok/s/user as the target on DeepSeek R1, H100 produces 154 tok/s/chip ($2.11 per million tokens) and H200 produces 500 ($0.68). H200 is 211% cheaper per token; H200 delivers 224% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H100:739.5H200:1772.7 | H100:280.6H200:907.0 | H100:154.3H200:500.1 |
| Cost ($/M tok) | H100:$0.439H200:$0.191 | H100:$1.158H200:$0.374 | H100:$2.106H200:$0.678 |
| tok/s/MW | H100:539778H200:1293964 | H100:204852H200:662067 | H100:112644H200:365052 |
| Concurrency | H100:~132H200:~115 | H100:~41H200:~122 | H100:~17H200:~4 |
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.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
DeepSeek R1 0528 671B 8K / 1K Total Tokens per $1 TCO vs. Interactivity