Muse Spark Open 816B — B300 vs Vera Rubin NVL72
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on Muse Spark Open 816B. 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 →
Setting 206 tok/s/user as the target on Muse Spark Open 816B, B300 produces 23299 tok/s/chip ($0.03 per million tokens) and Vera Rubin NVL72 produces 79984 ($0.01). Vera Rubin NVL72 is 115% cheaper per token; Vera Rubin NVL72 delivers 243% more tok/s/chip.
At 306 tok/s/user interactivity on Muse Spark Open 816B, B300 delivers 16082 tok/s/chip at $0.04 per million tokens; Vera Rubin NVL72 delivers 56067 tok/s/chip at $0.02. Vera Rubin NVL72 is 118% cheaper per token; Vera Rubin NVL72 delivers 249% more tok/s/chip at this point.
B300 posts 9002 tok/s/chip for $0.07 per million tokens at 405 tok/s/user on Muse Spark Open 816B; Vera Rubin NVL72 posts 42194 tok/s/chip for $0.02. Vera Rubin NVL72 is 193% cheaper per token; Vera Rubin NVL72 delivers 369% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:23298.7Vera Rubin NVL72:79984.1 | B300:16082.1Vera Rubin NVL72:56066.7 | B300:9001.8Vera Rubin NVL72:42194.0 |
| Cost ($/M tok) | B300:$0.027Vera Rubin NVL72:$0.013 | B300:$0.039Vera Rubin NVL72:$0.018 | B300:$0.070Vera Rubin NVL72:$0.024 |
| tok/s/MW | B300:12262465Vera Rubin NVL72:24237617 | B300:8464247Vera Rubin NVL72:16989904 | B300:4737794Vera Rubin NVL72:12786068 |
| Concurrency | B300:~17Vera Rubin NVL72:~45 | B300:~18Vera Rubin NVL72:~28 | B300:~11Vera Rubin NVL72:~32 |