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 — 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 →
Near the low end of the 3–347 tok/s/user interactivity band, at 89 tok/s/user on DeepSeek V4.1 Flash 552B: B300 runs 66335 tok/s/chip at $0.01/M tokens, GB300 NVL72 runs 10733 at $0.06/M. B300 is 532% cheaper per token; B300 delivers 518% more tok/s/chip.
Setting 175 tok/s/user as the target on DeepSeek V4.1 Flash 552B, B300 produces 27937 tok/s/chip ($0.02 per million tokens) and GB300 NVL72 produces 10733 ($0.06). B300 is 166% cheaper per token; B300 delivers 160% more tok/s/chip.
At 261 tok/s/user interactivity on DeepSeek V4.1 Flash 552B, B300 delivers 14248 tok/s/chip at $0.04 per million tokens; GB300 NVL72 delivers 10733 tok/s/chip at $0.06. B300 is 36% cheaper per token; B300 delivers 33% more tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:66334.6GB300 NVL72:10733.4 | B300:27937.4GB300 NVL72:10733.4 | B300:14248.2GB300 NVL72:10733.4 |
| Cost ($/M tok) | B300:$0.009GB300 NVL72:$0.060 | B300:$0.022GB300 NVL72:$0.060 | B300:$0.044GB300 NVL72:$0.060 |
| tok/s/MW | B300:34912934GB300 NVL72:5062944 | B300:14703918GB300 NVL72:5062944 | B300:7499077GB300 NVL72:5062944 |
| Concurrency | B300:~39GB300 NVL72:~16 | B300:~16GB300 NVL72:~16 | B300:~7GB300 NVL72:~16 |
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.
