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.
B200 posts 3161 tok/s/GPU for $0.17 per million tokens at 66 tok/s/user on Qwen 3.5 397B-A17B; GB200 NVL72 posts 1158 tok/s/GPU for $0.54. B200 is 214% cheaper per token; B200 delivers 173% more tok/s/GPU.
Throughput at 102 tok/s/user on Qwen 3.5 397B-A17B: B200 hits 1692 tok/s/GPU, GB200 NVL72 hits 456. Per-million costs land at $0.31 and $1.32 respectively. B200 is 322% cheaper per token; B200 delivers 271% more tok/s/GPU.
B200 / GB200 NVL72 on Qwen 3.5 397B-A17B at 138 tok/s/user: 972 / 201 tok/s/GPU, $0.56 / $3.06 per million tokens. B200 is 449% cheaper per token; B200 delivers 385% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:3161.2GB200 NVL72:1158.3 | B200:1691.9GB200 NVL72:455.6 | B200:971.8GB200 NVL72:200.6 |
| Cost ($/M tok) | B200:$0.172GB200 NVL72:$0.539 | B200:$0.313GB200 NVL72:$1.323 | B200:$0.557GB200 NVL72:$3.057 |
| tok/s/MW | B200:1456774GB200 NVL72:551548 | B200:779666GB200 NVL72:216965 | B200:447813GB200 NVL72:95505 |
| Concurrency | B200:~103GB200 NVL72:~88 | B200:~37GB200 NVL72:~19 | B200:~15GB200 NVL72:~6 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.