Qwen 3.5 397B-A17B — B200 vs MI355X Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) on Qwen 3.5 397B-A17B. Owning-hyperscaler TCO normalized by output tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
B200 edges MI355X at 82 tok/s/user on Qwen 3.5 397B-A17B — $0.22 per million tokens versus $0.26, a 19% cost-per-token gap.
Push Qwen 3.5 397B-A17B to 134 tok/s/user and B200 lands at $0.48 per million tokens against MI355X's $0.46 — MI355X pulls ahead by 6%.
B200: $0.75 per million tokens. MI355X: $0.78. Both at 185 tok/s/user on Qwen 3.5 397B-A17B, with B200 4% cheaper. (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.)
GPU pricing (owning hyperscaler): B200 $1.95/GPU/hr · MI355X $1.48/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B200:$0.219MI355X:$0.261 | B200:$0.484MI355X:$0.458 | B200:$0.751MI355X:$0.781 |
| Concurrency | B200:~68MI355X:~40 | B200:~19MI355X:~13 | B200:~9MI355X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.