Head-to-head AI inference benchmark comparison of B300 (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
Near the low end of the 52β826 tok/s/user interactivity band, at 245 tok/s/user on DeepSeek V4.1 Flash 552B: B300 runs 74566 tok/s/chip at $0.01/M tokens, GB300 NVL72 runs 526416 at $0.00/M. GB300 NVL72 is 591% cheaper per token; GB300 NVL72 delivers 606% more tok/s/chip.
Setting 439 tok/s/user as the target on DeepSeek V4.1 Flash 552B, B300 produces 33439 tok/s/chip ($0.02 per million tokens) and GB300 NVL72 produces 29405 ($0.02). B300 is 16% cheaper per token; B300 delivers 14% more tok/s/chip.
At 633 tok/s/user interactivity on DeepSeek V4.1 Flash 552B, B300 delivers 13715 tok/s/chip at $0.05 per million tokens; GB300 NVL72 delivers 9832 tok/s/chip at $0.07. B300 is 43% cheaper per token; B300 delivers 39% more tok/s/chip at this point. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:74565.9GB300 NVL72:526415.9 | B300:33439.4GB300 NVL72:29405.2 | B300:13714.9GB300 NVL72:9832.0 |
| Cost ($/M tok) | B300:$0.008GB300 NVL72:$0.001 | B300:$0.019GB300 NVL72:$0.022 | B300:$0.046GB300 NVL72:$0.065 |
| tok/s/MW | B300:39245191GB300 NVL72:248309393 | B300:17599688GB300 NVL72:13870373 | B300:7218350GB300 NVL72:4637728 |
| Concurrency | B300:~22GB300 NVL72:~2923 | B300:~19GB300 NVL72:~16 | B300:~5GB300 NVL72:~3 |
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.
