Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Kimi K3 2.8T. 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 →
B200 posts 4173 tok/s/chip for $0.12 per million tokens at 56 tok/s/user on Kimi K3 2.8T; GB200 NVL72 posts 4090 tok/s/chip for $0.13. B200 is 10% cheaper per token; B200 delivers 2% more tok/s/chip.
Throughput at 104 tok/s/user on Kimi K3 2.8T: B200 hits 2102 tok/s/chip, GB200 NVL72 hits 2269. Per-million costs land at $0.23 and $0.23 respectively. Cost per token is essentially tied; GB200 NVL72 delivers 8% more tok/s/chip.
B200 / GB200 NVL72 on Kimi K3 2.8T at 151 tok/s/user: 1311 / 1020 tok/s/chip, $0.37 / $0.51 per million tokens. B200 is 38% cheaper per token; B200 delivers 29% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:4173.1GB200 NVL72:4089.6 | B200:2102.0GB200 NVL72:2268.8 | B200:1310.8GB200 NVL72:1019.9 |
| Cost ($/M tok) | B200:$0.115GB200 NVL72:$0.126 | B200:$0.229GB200 NVL72:$0.228 | B200:$0.367GB200 NVL72:$0.507 |
| tok/s/MW | B200:2440388GB200 NVL72:2186971 | B200:1229257GB200 NVL72:1213257 | B200:766551GB200 NVL72:545383 |
| Concurrency | B200:~15GB200 NVL72:~16 | B200:~6GB200 NVL72:~16 | B200:~3GB200 NVL72:~2 |
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.
