Qwen 3.5 397B-A17B — MI325X vs MI355X Performance per Dollar
Cost per million tokens of MI325X (AMD CDNA 3) 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.
MI325X: $0.74 per million tokens. MI355X: $0.15. Both at 46 tok/s/user on Qwen 3.5 397B-A17B, with MI355X 380% cheaper.
Around the middle of the 37–72 tok/s/user interactivity band — at 55 tok/s/user — MI325X runs $1.40 per million tokens on Qwen 3.5 397B-A17B while MI355X runs $0.17. MI355X is the cheaper choice by 710%.
On Qwen 3.5 397B-A17B at 64 tok/s/user, the per-million math comes out to $2.27 for MI325X and $0.20 for MI355X; MI355X delivers 1062% 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.)
Chip pricing (owning hyperscaler): MI325X $1.10/chip/hr · MI355X $1.50/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 | MI325X:$0.743MI355X:$0.155 | MI325X:$1.402MI355X:$0.173 | MI325X:$2.267MI355X:$0.195 |
| Concurrency | MI325X:~37MI355X:~116 | MI325X:~16MI355X:~85 | MI325X:~9MI355X:~68 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.