Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB300 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
Setting 202 tok/s/user as the target on DeepSeek V4.1 Flash 552B, B200 produces 163340 tok/s/chip ($0.00 per million tokens) and GB300 NVL72 produces 945219 ($0.00). GB300 NVL72 is 333% cheaper per token; GB300 NVL72 delivers 479% more tok/s/chip.
At 352 tok/s/user interactivity on DeepSeek V4.1 Flash 552B, B200 delivers 55132 tok/s/chip at $0.01 per million tokens; GB300 NVL72 delivers 280054 tok/s/chip at $0.00. GB300 NVL72 is 280% cheaper per token; GB300 NVL72 delivers 408% more tok/s/chip at this point.
B200 posts 18312 tok/s/chip for $0.03 per million tokens at 503 tok/s/user on DeepSeek V4.1 Flash 552B; GB300 NVL72 posts 9787 tok/s/chip for $0.07. B200 is 150% cheaper per token; B200 delivers 87% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:163339.7GB300 NVL72:945219.3 | B200:55132.0GB300 NVL72:280053.8 | B200:18312.4GB300 NVL72:9787.5 |
| Cost ($/M tok) | B200:$0.003GB300 NVL72:$0.001 | B200:$0.009GB300 NVL72:$0.002 | B200:$0.026GB300 NVL72:$0.066 |
| tok/s/MW | B200:95520318GB300 NVL72:445858151 | B200:32240958GB300 NVL72:132100833 | B200:10708996GB300 NVL72:4616737 |
| Concurrency | B200:~100GB300 NVL72:~10861 | B200:~30GB300 NVL72:~1671 | B200:~9GB300 NVL72:~4 |
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.
