Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) 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.
Throughput at 90 tok/s/user on gpt-oss 120B: H100 hits 6362 tok/s/chip, H200 hits 6798. Per-million costs land at $0.05 and $0.05 respectively. H200 is 2% cheaper per token; H200 delivers 7% more tok/s/chip.
H100 / H200 on gpt-oss 120B at 143 tok/s/user: 4148 / 4618 tok/s/chip, $0.08 / $0.07 per million tokens. H200 is 7% cheaper per token; H200 delivers 11% more tok/s/chip.
Toward the upper edge of the 37–250 tok/s/user interactivity band, at 197 tok/s/user on gpt-oss 120B: H100 runs 2636 tok/s/chip at $0.12/M tokens, H200 runs 2945 at $0.12/M. H200 is 7% cheaper per token; H200 delivers 12% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H100:6361.9H200:6797.8 | H100:4147.7H200:4618.1 | H100:2636.0H200:2945.5 |
| Cost ($/M tok) | H100:$0.051H200:$0.050 | H100:$0.078H200:$0.073 | H100:$0.123H200:$0.115 |
| tok/s/MW | H100:4643748H200:4961923 | H100:3027501H200:3370852 | H100:1924056H200:2149964 |
| Concurrency | H100:~32H200:~19 | H100:~13H200:~16 | H100:~6H200:~7 |
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 • 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