Qwen 3.5 397B-A17B · Performance per Dollar

Qwen 3.5 397B-A17B — H200 vs MI300X Performance per Dollar

Cost per million tokens of H200 (NVIDIA Hopper) 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.

H200 edges MI300X at 43 tok/s/user on Qwen 3.5 397B-A17B — $0.70 per million tokens versus $0.89, a 27% cost-per-token gap.

Push Qwen 3.5 397B-A17B to 53 tok/s/user and H200 lands at $0.82 per million tokens against MI300X's $1.63 — H200 pulls ahead by 99%.

H200: $0.94 per million tokens. MI300X: $2.52. Both at 62 tok/s/user on Qwen 3.5 397B-A17B, with H200 167% cheaper. (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 · MI300X $1.12/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
H200:$0.703MI300X:$0.893
H200:$0.821MI300X:$1.633
H200:$0.943MI300X:$2.517
Concurrency
H200:~51MI300X:~34
H200:~37MI300X:~15
H200:~27MI300X:~8

Inference Performance

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