Qwen 3.5 397B-A17B — GB200 NVL72 vs MI300X
Head-to-head AI inference benchmark comparison of GB200 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.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX →
GB200 NVL72 hits 52727 tok/s/chip for $0.01 per million tokens at 168 tok/s/user on Qwen 3.5 397B-A17B. No MI300X data at this operating point.
GB200 NVL72: 24309 tok/s/chip, $0.02 per million tokens at 287 tok/s/user on Qwen 3.5 397B-A17B. MI300X is unmeasured here.
At 405 tok/s/user on Qwen 3.5 397B-A17B, GB200 NVL72 delivers 9186 tok/s/chip at $0.06 per million tokens; MI300X hasn't been benchmarked at this target. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:52727.4MI300X:— | GB200 NVL72:24309.2MI300X:— | GB200 NVL72:9185.8MI300X:— |
| Cost ($/M tok) | GB200 NVL72:$0.010MI300X:— | GB200 NVL72:$0.021MI300X:— | GB200 NVL72:$0.056MI300X:— |
| tok/s/MW | GB200 NVL72:28196482MI300X:— | GB200 NVL72:12999597MI300X:— | GB200 NVL72:4912176MI300X:— |
| Concurrency | GB200 NVL72:~23MI300X:— | GB200 NVL72:~15MI300X:— | GB200 NVL72:~6MI300X:— |