Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Muse Spark Open 816B. 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
B200 / GB200 NVL72 on Muse Spark Open 816B at 143 tok/s/user: 16452 / 16506 tok/s/chip, $0.03 / $0.03 per million tokens. B200 is 7% cheaper per token; throughput per chip is essentially tied.
Around the middle of the 84β321 tok/s/user interactivity band, at 203 tok/s/user on Muse Spark Open 816B: B200 runs 7581 tok/s/chip at $0.06/M tokens, GB200 NVL72 runs 7382 at $0.07/M. B200 is 10% cheaper per token; B200 delivers 3% more tok/s/chip.
Setting 262 tok/s/user as the target on Muse Spark Open 816B, B200 produces 5276 tok/s/chip ($0.09 per million tokens) and GB200 NVL72 produces 5594 ($0.09). B200 is 1% cheaper per token; GB200 NVL72 delivers 6% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:16451.7GB200 NVL72:16506.4 | B200:7581.4GB200 NVL72:7381.6 | B200:5275.8GB200 NVL72:5594.4 |
| Cost ($/M tok) | B200:$0.029GB200 NVL72:$0.031 | B200:$0.063GB200 NVL72:$0.070 | B200:$0.091GB200 NVL72:$0.092 |
| tok/s/MW | B200:9620898GB200 NVL72:8826943 | B200:4433570GB200 NVL72:3947395 | B200:3085237GB200 NVL72:2991641 |
| Concurrency | B200:~23GB200 NVL72:~12 | B200:~10GB200 NVL72:~5 | B200:~8GB200 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.
