Qwen 3.5 397B-A17B · Performance per Dollar

Qwen 3.5 397B-A17B — MI300X vs MI325X Performance per Dollar

Cost per million tokens of MI300X (AMD CDNA 3) versus MI325X (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.

On Qwen 3.5 397B-A17B at 55 tok/s/user, the per-million math comes out to $0.99 for MI300X and $0.89 for MI325X; MI325X delivers 11% more output per dollar.

At 68 tok/s/user on Qwen 3.5 397B-A17B, MI300X costs $1.37 per million tokens; MI325X costs $1.25. MI325X is 10% more cost-efficient at this operating point.

MI325X edges MI300X at 82 tok/s/user on Qwen 3.5 397B-A17B — $1.93 per million tokens versus $2.06, a 7% cost-per-token gap. (Numbers reflect the default 1k/1k · bf16 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

GPU pricing (owning hyperscaler): MI300X $1.12/GPU/hr · MI325X $1.28/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.

View full latency + throughput comparison →

Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
Metric
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Dollar per Million Tokens
MI300X:$0.993MI325X:$0.891
MI300X:$1.375MI325X:$1.252
MI300X:$2.062MI325X:$1.926
Concurrency
MI300X:~23MI325X:~32
MI300X:~13MI325X:~17
MI300X:~8MI325X:~10

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.