Qwen 3.5 397B-A17B — GB200 NVL72 vs MI355X Performance per Dollar
Cost per million tokens of GB200 NVL72 (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.
MI355X edges GB200 NVL72 at 64 tok/s/user on Qwen 3.5 397B-A17B — $0.19 per million tokens versus $0.51, a 163% cost-per-token gap.
Push Qwen 3.5 397B-A17B to 101 tok/s/user and GB200 NVL72 lands at $1.30 per million tokens against MI355X's $0.33 — MI355X pulls ahead by 293%.
GB200 NVL72: $2.98 per million tokens. MI355X: $0.47. Both at 137 tok/s/user on Qwen 3.5 397B-A17B, with MI355X 529% 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): GB200 NVL72 $2.21/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 | GB200 NVL72:$0.506MI355X:$0.193 | GB200 NVL72:$1.297MI355X:$0.330 | GB200 NVL72:$2.983MI355X:$0.475 |
| Concurrency | GB200 NVL72:~113MI355X:~68 | GB200 NVL72:~20MI355X:~25 | GB200 NVL72:~7MI355X:~13 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.