Qwen 3.5 397B-A17B · GPU comparison

Qwen 3.5 397B-A17B — H100 vs MI300X

Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) and MI300X (AMD CDNA 3) on Qwen 3.5 397B-A17B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.

H100 posts 308 tok/s/GPU for $1.20 per million tokens at 52 tok/s/user on Qwen 3.5 397B-A17B; MI300X posts 201 tok/s/GPU for $1.55. H100 is 29% cheaper per token; H100 delivers 53% more tok/s/GPU.

Throughput at 58 tok/s/user on Qwen 3.5 397B-A17B: H100 hits 270 tok/s/GPU, MI300X hits 147. Per-million costs land at $1.35 and $2.04 respectively. H100 is 51% cheaper per token; H100 delivers 84% more tok/s/GPU.

H100 / MI300X on Qwen 3.5 397B-A17B at 65 tok/s/user: 235 / 104 tok/s/GPU, $1.53 / $3.05 per million tokens. H100 is 100% cheaper per token; H100 delivers 126% more tok/s/GPU. (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.)

View performance-per-dollar view →

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)
Throughput (tok/s/gpu)
H100:307.6MI300X:201.4
H100:270.4MI300X:147.3
H100:234.9MI300X:103.9
Cost ($/M tok)
H100:$1.198MI300X:$1.547
H100:$1.352MI300X:$2.044
H100:$1.527MI300X:$3.054
tok/s/MW
H100:177792MI300X:112516
H100:156311MI300X:82288
H100:135793MI300X:58024
Concurrency
H100:~25MI300X:~16
H100:~19MI300X:~11
H100:~15MI300X:~7

Inference Performance

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