Qwen 3.5 397B-A17B — B200 vs MI325X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI325X (AMD CDNA 3) 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 / MI325X on Qwen 3.5 397B-A17B at 46 tok/s/user: 4415 / 403 tok/s/GPU, $0.12 / $0.86 per million tokens. B200 is 601% cheaper per token; B200 delivers 996% more tok/s/GPU.
Around the middle of the 37–72 tok/s/user interactivity band, at 55 tok/s/user on Qwen 3.5 397B-A17B: B200 runs 3811 tok/s/GPU at $0.14/M tokens, MI325X runs 219 at $1.63/M. B200 is 1047% cheaper per token; B200 delivers 1637% more tok/s/GPU.
Setting 64 tok/s/user as the target on Qwen 3.5 397B-A17B, B200 produces 3271 tok/s/GPU ($0.17 per million tokens) and MI325X produces 134 ($2.64). B200 is 1491% cheaper per token; B200 delivers 2337% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:4415.5MI325X:402.9 | B200:3810.6MI325X:219.4 | B200:3271.4MI325X:134.3 |
| Cost ($/M tok) | B200:$0.123MI325X:$0.864 | B200:$0.142MI325X:$1.631 | B200:$0.166MI325X:$2.638 |
| tok/s/MW | B200:2034784MI325X:184830 | B200:1756045MI325X:100633 | B200:1507578MI325X:61591 |
| Concurrency | B200:~215MI325X:~37 | B200:~155MI325X:~16 | B200:~110MI325X:~9 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.