Qwen 3.5 397B-A17B — GB200 NVL72 vs MI300X Performance per Dollar
Cost per million tokens of GB200 NVL72 (NVIDIA Blackwell) versus MI300X (AMD CDNA 3) 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.
On Qwen 3.5 397B-A17B at 43 tok/s/user, the per-million math comes out to $0.04 for GB200 NVL72 and $0.34 for MI300X; GB200 NVL72 delivers 745% more output per dollar.
At 51 tok/s/user on Qwen 3.5 397B-A17B, GB200 NVL72 costs $0.04 per million tokens; MI300X costs $0.47. GB200 NVL72 is 1017% more cost-efficient at this operating point.
GB200 NVL72 edges MI300X at 59 tok/s/user on Qwen 3.5 397B-A17B — $0.04 per million tokens versus $0.68, a 1449% cost-per-token gap. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Chip pricing (owning hyperscaler): GB200 NVL72 $1.86/chip/hr · MI300X $0.95/chip/hr. Source: SemiAnalysis Market July 2026 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:$0.041MI300X:$0.344 | GB200 NVL72:$0.042MI300X:$0.469 | GB200 NVL72:$0.044MI300X:$0.682 |
| Concurrency | GB200 NVL72:~2304MI300X:~17 | GB200 NVL72:~1975MI300X:~11 | GB200 NVL72:~1447MI300X:~6 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.