Qwen 3.5 397B-A17B — H200 vs MI355X Performance per Dollar
Cost per million tokens of H200 (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.
H200: $0.82 per million tokens. MI355X: $0.17. Both at 53 tok/s/user on Qwen 3.5 397B-A17B, with MI355X 393% cheaper.
Around the middle of the 27–132 tok/s/user interactivity band — at 80 tok/s/user — H200 runs $1.22 per million tokens on Qwen 3.5 397B-A17B while MI355X runs $0.25. MI355X is the cheaper choice by 381%.
On Qwen 3.5 397B-A17B at 106 tok/s/user, the per-million math comes out to $1.44 for H200 and $0.35 for MI355X; MI355X delivers 317% 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): H200 $1.41/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 | H200:$0.821MI355X:$0.167 | H200:$1.216MI355X:$0.253 | H200:$1.438MI355X:$0.345 |
| Concurrency | H200:~37MI355X:~90 | H200:~17MI355X:~42 | H200:~12MI355X:~22 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.