Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on DeepSeek V4.1 Flash 552B. 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
Throughput at 183 tok/s/user on DeepSeek V4.1 Flash 552B: B200 hits 72057 tok/s/chip, GB200 NVL72 hits 81711. Per-million costs land at $0.01 and $0.01 respectively. GB200 NVL72 is 5% cheaper per token; GB200 NVL72 delivers 13% more tok/s/chip.
B200 / GB200 NVL72 on DeepSeek V4.1 Flash 552B at 294 tok/s/user: 30525 / 16595 tok/s/chip, $0.02 / $0.03 per million tokens. B200 is 98% cheaper per token; B200 delivers 84% more tok/s/chip.
Toward the upper edge of the 73β515 tok/s/user interactivity band, at 405 tok/s/user on DeepSeek V4.1 Flash 552B: B200 runs 16041 tok/s/chip at $0.03/M tokens, GB200 NVL72 runs 15649 at $0.03/M. B200 is 10% cheaper per token; B200 delivers 3% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:72057.2GB200 NVL72:81710.5 | B200:30524.7GB200 NVL72:16595.1 | B200:16041.1GB200 NVL72:15649.1 |
| Cost ($/M tok) | B200:$0.007GB200 NVL72:$0.006 | B200:$0.016GB200 NVL72:$0.031 | B200:$0.030GB200 NVL72:$0.033 |
| tok/s/MW | B200:42138724GB200 NVL72:43695479 | B200:17850729GB200 NVL72:8874397 | B200:9380757GB200 NVL72:8368499 |
| Concurrency | B200:~47GB200 NVL72:~42 | B200:~9GB200 NVL72:~4 | B200:~8GB200 NVL72:~8 |
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.
