Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on DeepSeekv4 Pro 0813 1.6T. 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
B300 posts 21845 tok/s/chip for $0.03 per million tokens at 129 tok/s/user on DeepSeekv4 Pro 0813 1.6T; Vera Rubin NVL72 posts 256644 tok/s/chip for $0.00. Vera Rubin NVL72 is 635% cheaper per token; Vera Rubin NVL72 delivers 1075% more tok/s/chip.
Throughput at 185 tok/s/user on DeepSeekv4 Pro 0813 1.6T: B300 hits 9447 tok/s/chip, Vera Rubin NVL72 hits 142100. Per-million costs land at $0.07 and $0.01 respectively. Vera Rubin NVL72 is 842% cheaper per token; Vera Rubin NVL72 delivers 1404% more tok/s/chip.
B300 / Vera Rubin NVL72 on DeepSeekv4 Pro 0813 1.6T at 240 tok/s/user: 3240 / 4903 tok/s/chip, $0.19 / $0.20 per million tokens. B300 is 6% cheaper per token; Vera Rubin NVL72 delivers 51% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:21845.0Vera Rubin NVL72:256643.8 | B300:9447.0Vera Rubin NVL72:142100.1 | B300:3240.3Vera Rubin NVL72:4903.3 |
| Cost ($/M tok) | B300:$0.029Vera Rubin NVL72:$0.004 | B300:$0.066Vera Rubin NVL72:$0.007 | B300:$0.194Vera Rubin NVL72:$0.205 |
| tok/s/MW | B300:11497379Vera Rubin NVL72:77770858 | B300:4972098Vera Rubin NVL72:43060633 | B300:1705441Vera Rubin NVL72:1485860 |
| Concurrency | B300:~64Vera Rubin NVL72:~2749 | B300:~4Vera Rubin NVL72:~1288 | B300:~4Vera Rubin NVL72:~21 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
