Qwen 3.5 397B-A17B — GB300 NVL72 vs MI300X
Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) 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.
GB300 NVL72 hits 6025 tok/s/GPU for $0.12 per million tokens at 90 tok/s/user on Qwen 3.5 397B-A17B. No MI300X data at this operating point.
GB300 NVL72: 2801 tok/s/GPU, $0.26 per million tokens at 136 tok/s/user on Qwen 3.5 397B-A17B. MI300X is unmeasured here.
At 183 tok/s/user on Qwen 3.5 397B-A17B, GB300 NVL72 delivers 1338 tok/s/GPU at $0.55 per million tokens; MI300X hasn't been benchmarked at this target. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | GB300 NVL72:6024.7MI300X:— | GB300 NVL72:2801.0MI300X:— | GB300 NVL72:1338.4MI300X:— |
| Cost ($/M tok) | GB300 NVL72:$0.122MI300X:— | GB300 NVL72:$0.256MI300X:— | GB300 NVL72:$0.553MI300X:— |
| tok/s/MW | GB300 NVL72:2868920MI300X:— | GB300 NVL72:1333831MI300X:— | GB300 NVL72:637343MI300X:— |
| Concurrency | GB300 NVL72:~71MI300X:— | GB300 NVL72:~21MI300X:— | GB300 NVL72:~7MI300X:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.