Kimi K3.1 Pro — B200 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Kimi K3.1 Pro. 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 / GB200 NVL72 on Kimi K3.1 Pro at 40 tok/s/user: 2761 / 8792 tok/s/chip, $0.17 / $0.06 per million tokens. GB200 NVL72 is 196% cheaper per token; GB200 NVL72 delivers 218% more tok/s/chip.
Around the middle of the 27–83 tok/s/user interactivity band, at 55 tok/s/user on Kimi K3.1 Pro: B200 runs 1838 tok/s/chip at $0.26/M tokens, GB200 NVL72 runs 8445 at $0.06/M. GB200 NVL72 is 327% cheaper per token; GB200 NVL72 delivers 359% more tok/s/chip.
Setting 69 tok/s/user as the target on Kimi K3.1 Pro, B200 produces 1544 tok/s/chip ($0.31 per million tokens) and GB200 NVL72 produces 8272 ($0.06). GB200 NVL72 is 398% cheaper per token; GB200 NVL72 delivers 436% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
View performance-per-dollar view →
GB200 NVL72 vs B200: full specs & benchmarks across models →
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:2761.1GB200 NVL72:8792.4 | B200:1837.9GB200 NVL72:8444.7 | B200:1543.7GB200 NVL72:8272.2 |
| Cost ($/M tok) | B200:$0.174GB200 NVL72:$0.059 | B200:$0.261GB200 NVL72:$0.061 | B200:$0.311GB200 NVL72:$0.062 |
| tok/s/MW | B200:1614678GB200 NVL72:4701839 | B200:1074770GB200 NVL72:4515904 | B200:902756GB200 NVL72:4423632 |
| Concurrency | B200:~13GB200 NVL72:~32 | B200:~7GB200 NVL72:~37 | B200:~5GB200 NVL72:~44 |