Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on Qwen 3.8 Flash Next 176B-A6B. 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.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX β
H100: 4980 tok/s/chip, $0.07 per million tokens at 46 tok/s/user on Qwen 3.8 Flash Next 176B-A6B. Vera Rubin NVL72 is unmeasured here.
At 77 tok/s/user on Qwen 3.8 Flash Next 176B-A6B, H100 delivers 4944 tok/s/chip at $0.07 per million tokens; Vera Rubin NVL72 hasn't been benchmarked at this target.
H100 hits 4908 tok/s/chip for $0.07 per million tokens at 108 tok/s/user on Qwen 3.8 Flash Next 176B-A6B. No Vera Rubin NVL72 data at this operating point. (Numbers reflect the default agentic-traces Β· fp8 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H100:4979.7Vera Rubin NVL72:β | H100:4944.2Vera Rubin NVL72:β | H100:4907.6Vera Rubin NVL72:β |
| Cost ($/M tok) | H100:$0.065Vera Rubin NVL72:β | H100:$0.066Vera Rubin NVL72:β | H100:$0.066Vera Rubin NVL72:β |
| tok/s/MW | H100:3634841Vera Rubin NVL72:β | H100:3608888Vera Rubin NVL72:β | H100:3582201Vera Rubin NVL72:β |
| Concurrency | H100:~11Vera Rubin NVL72:β | H100:~10Vera Rubin NVL72:β | H100:~9Vera Rubin NVL72:β |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
