Qwen 3.5 397B-A17B — H100 vs MI355X
Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) and MI355X (AMD CDNA 4) 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.
Setting 63 tok/s/user as the target on Qwen 3.5 397B-A17B, H100 produces 517 tok/s/GPU ($0.73 per million tokens) and MI355X produces 2160 ($0.19). MI355X is 283% cheaper per token; MI355X delivers 318% more tok/s/GPU.
At 97 tok/s/user interactivity on Qwen 3.5 397B-A17B, H100 delivers 220 tok/s/GPU at $1.66 per million tokens; MI355X delivers 1273 tok/s/GPU at $0.32. MI355X is 424% cheaper per token; MI355X delivers 479% more tok/s/GPU at this point.
H100 posts 86 tok/s/GPU for $4.06 per million tokens at 131 tok/s/user on Qwen 3.5 397B-A17B; MI355X posts 933 tok/s/GPU for $0.44. MI355X is 818% cheaper per token; MI355X delivers 983% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | H100:517.2MI355X:2160.5 | H100:220.0MI355X:1273.0 | H100:86.2MI355X:933.5 |
| Cost ($/M tok) | H100:$0.726MI355X:$0.190 | H100:$1.664MI355X:$0.318 | H100:$4.057MI355X:$0.442 |
| tok/s/MW | H100:298955MI355X:815281 | H100:127176MI355X:480378 | H100:49829MI355X:352251 |
| Concurrency | H100:~39MI355X:~69 | H100:~10MI355X:~27 | H100:~3MI355X:~14 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.