Qwen 3.5 397B-A17B — B300 vs MI300X Performance per Dollar
Cost per million tokens of B300 (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.
Near the low end of the 34–71 tok/s/user interactivity band — at 43 tok/s/user — B300 runs $0.14 per million tokens on Qwen 3.5 397B-A17B while MI300X runs $0.89. B300 is the cheaper choice by 528%.
On Qwen 3.5 397B-A17B at 53 tok/s/user, the per-million math comes out to $0.17 for B300 and $1.63 for MI300X; B300 delivers 846% more output per dollar.
At 62 tok/s/user on Qwen 3.5 397B-A17B, B300 costs $0.20 per million tokens; MI300X costs $2.52. B300 is 1149% more cost-efficient at this operating point. (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): B300 $2.34/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 | B300:$0.142MI300X:$0.893 | B300:$0.173MI300X:$1.633 | B300:$0.201MI300X:$2.517 |
| Concurrency | B300:~230MI300X:~34 | B300:~153MI300X:~15 | B300:~106MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.