Qwen 3.5 397B-A17B — H100 vs MI355X Performance per Dollar
Cost per million tokens of H100 (NVIDIA Hopper) 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.
Near the low end of the 29–171 tok/s/user interactivity band — at 64 tok/s/user — H100 runs $0.85 per million tokens on Qwen 3.5 397B-A17B while MI355X runs $0.19. MI355X is the cheaper choice by 340%.
On Qwen 3.5 397B-A17B at 100 tok/s/user, the per-million math comes out to $1.15 for H100 and $0.33 for MI355X; MI355X delivers 252% more output per dollar.
At 135 tok/s/user on Qwen 3.5 397B-A17B, H100 costs $1.54 per million tokens; MI355X costs $0.46. MI355X is 232% 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): H100 $1.30/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 | H100:$0.848MI355X:$0.193 | H100:$1.151MI355X:$0.327 | H100:$1.536MI355X:$0.463 |
| Concurrency | H100:~28MI355X:~68 | H100:~13MI355X:~25 | H100:~7MI355X:~13 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.