Qwen 3.5 397B-A17B — B200 vs MI300X Performance per Dollar
Cost per million tokens of B200 (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.
B200: $0.12 per million tokens. MI300X: $0.89. Both at 43 tok/s/user on Qwen 3.5 397B-A17B, with B200 660% cheaper.
Around the middle of the 34–71 tok/s/user interactivity band — at 53 tok/s/user — B200 runs $0.14 per million tokens on Qwen 3.5 397B-A17B while MI300X runs $1.63. B200 is the cheaper choice by 1086%.
On Qwen 3.5 397B-A17B at 62 tok/s/user, the per-million math comes out to $0.16 for B200 and $2.52 for MI300X; B200 delivers 1471% more output per dollar. (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 · MI300X $1.12/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:$0.118MI300X:$0.893 | B200:$0.138MI300X:$1.633 | B200:$0.160MI300X:$2.517 |
| Concurrency | B200:~237MI300X:~34 | B200:~167MI300X:~15 | B200:~118MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.