Muse Spark Open 816B — GB200 NVL72 vs Vera Rubin NVL72
Head-to-head AI inference benchmark comparison of GB200 NVL72 (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 →
Near the low end of the 107–377 tok/s/user interactivity band, at 174 tok/s/user on Muse Spark Open 816B: GB200 NVL72 runs 20043 tok/s/chip at $0.03/M tokens, Vera Rubin NVL72 runs 94020 at $0.01/M. Vera Rubin NVL72 is 142% cheaper per token; Vera Rubin NVL72 delivers 369% more tok/s/chip.
Setting 242 tok/s/user as the target on Muse Spark Open 816B, GB200 NVL72 produces 14453 tok/s/chip ($0.04 per million tokens) and Vera Rubin NVL72 produces 68769 ($0.01). Vera Rubin NVL72 is 145% cheaper per token; Vera Rubin NVL72 delivers 376% more tok/s/chip.
At 310 tok/s/user interactivity on Muse Spark Open 816B, GB200 NVL72 delivers 8523 tok/s/chip at $0.06 per million tokens; Vera Rubin NVL72 delivers 55469 tok/s/chip at $0.02. Vera Rubin NVL72 is 235% cheaper per token; Vera Rubin NVL72 delivers 551% more tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:20043.5Vera Rubin NVL72:94020.2 | GB200 NVL72:14452.9Vera Rubin NVL72:68768.5 | GB200 NVL72:8522.7Vera Rubin NVL72:55469.2 |
| Cost ($/M tok) | GB200 NVL72:$0.026Vera Rubin NVL72:$0.011 | GB200 NVL72:$0.036Vera Rubin NVL72:$0.015 | GB200 NVL72:$0.061Vera Rubin NVL72:$0.018 |
| tok/s/MW | GB200 NVL72:10718434Vera Rubin NVL72:28490956 | GB200 NVL72:7728841Vera Rubin NVL72:20838944 | GB200 NVL72:4557599Vera Rubin NVL72:16808839 |
| Concurrency | GB200 NVL72:~12Vera Rubin NVL72:~54 | GB200 NVL72:~9Vera Rubin NVL72:~38 | GB200 NVL72:~4Vera Rubin NVL72:~28 |