Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) 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
B200 / Vera Rubin NVL72 on Qwen 3.8 Flash Next 176B-A6B at 104 tok/s/user: 56330 / 208116 tok/s/chip, $0.01 / $0.00 per million tokens. Vera Rubin NVL72 is 77% cheaper per token; Vera Rubin NVL72 delivers 269% more tok/s/chip.
Around the middle of the 53β259 tok/s/user interactivity band, at 156 tok/s/user on Qwen 3.8 Flash Next 176B-A6B: B200 runs 35280 tok/s/chip at $0.01/M tokens, Vera Rubin NVL72 runs 159188 at $0.01/M. Vera Rubin NVL72 is 116% cheaper per token; Vera Rubin NVL72 delivers 351% more tok/s/chip.
Setting 207 tok/s/user as the target on Qwen 3.8 Flash Next 176B-A6B, B200 produces 18616 tok/s/chip ($0.03 per million tokens) and Vera Rubin NVL72 produces 104162 ($0.01). Vera Rubin NVL72 is 168% cheaper per token; Vera Rubin NVL72 delivers 460% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:56330.0Vera Rubin NVL72:208115.7 | B200:35279.9Vera Rubin NVL72:159187.8 | B200:18616.5Vera Rubin NVL72:104162.4 |
| Cost ($/M tok) | B200:$0.009Vera Rubin NVL72:$0.005 | B200:$0.014Vera Rubin NVL72:$0.006 | B200:$0.026Vera Rubin NVL72:$0.010 |
| tok/s/MW | B200:32941517Vera Rubin NVL72:63065349 | B200:20631535Vera Rubin NVL72:48238716 | B200:10886821Vera Rubin NVL72:31564372 |
| Concurrency | B200:~12Vera Rubin NVL72:~272 | B200:~7Vera Rubin NVL72:~265 | B200:~2Vera Rubin NVL72:~173 |
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.
