Qwen 3.5 397B-A17B — GB300 NVL72 vs MI300X Performance per Dollar
Cost per million tokens of GB300 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.
Push Qwen 3.5 397B-A17B to 46 tok/s/user and GB300 NVL72 lands at $0.05 per million tokens against MI300X's $0.39 — GB300 NVL72 pulls ahead by 693%.
GB300 NVL72: $0.05 per million tokens. MI300X: $0.51. Both at 53 tok/s/user on Qwen 3.5 397B-A17B, with GB300 NVL72 945% cheaper.
Toward the upper edge of the 40–66 tok/s/user interactivity band — at 60 tok/s/user — GB300 NVL72 runs $0.05 per million tokens on Qwen 3.5 397B-A17B while MI300X runs $0.72. GB300 NVL72 is the cheaper choice by 1314%. (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): GB300 NVL72 $2.31/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 | GB300 NVL72:$0.049MI300X:$0.387 | GB300 NVL72:$0.049MI300X:$0.510 | GB300 NVL72:$0.051MI300X:$0.718 |
| Concurrency | GB300 NVL72:~2304MI300X:~14 | GB300 NVL72:~2304MI300X:~9 | GB300 NVL72:~1708MI300X:~6 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.