Qwen 3.5 397B-A17B — B200 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Qwen 3.5 397B-A17B. 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 56131 tok/s/chip for $0.01 per million tokens at 135 tok/s/user on Qwen 3.5 397B-A17B; GB200 NVL72 posts 65931 tok/s/chip for $0.01. GB200 NVL72 is 9% cheaper per token; GB200 NVL72 delivers 17% more tok/s/chip.
Throughput at 220 tok/s/user on Qwen 3.5 397B-A17B: B200 hits 32347 tok/s/chip, GB200 NVL72 hits 33148. Per-million costs land at $0.01 and $0.02 respectively. B200 is 5% cheaper per token; GB200 NVL72 delivers 2% more tok/s/chip.
B200 / GB200 NVL72 on Qwen 3.5 397B-A17B at 305 tok/s/user: 13596 / 22083 tok/s/chip, $0.04 / $0.02 per million tokens. GB200 NVL72 is 51% cheaper per token; GB200 NVL72 delivers 62% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
View performance-per-dollar view →
GB200 NVL72 vs B200: full specs & benchmarks across models →
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:56131.2GB200 NVL72:65931.3 | B200:32347.0GB200 NVL72:33147.8 | B200:13595.9GB200 NVL72:22082.8 |
| Cost ($/M tok) | B200:$0.009GB200 NVL72:$0.008 | B200:$0.015GB200 NVL72:$0.016 | B200:$0.035GB200 NVL72:$0.023 |
| tok/s/MW | B200:32825234GB200 NVL72:35257386 | B200:18916372GB200 NVL72:17726096 | B200:7950813GB200 NVL72:11809006 |
| Concurrency | B200:~21GB200 NVL72:~22 | B200:~22GB200 NVL72:~23 | B200:~10GB200 NVL72:~13 |