Qwen 3.5 397B-A17B · Chip comparison
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.
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 →
GB300 NVL72 posts 57,478 tok/s/chip for $0.011 per million tokens at 148 tok/s/user on Qwen 3.5 397B-A17B; MI300X posts 713 tok/s/chip for $0.37. GB300 NVL72 is 3217% cheaper per token; GB300 NVL72 delivers 7966% more tok/s/chip.
Throughput at 292 tok/s/user on Qwen 3.5 397B-A17B: GB300 NVL72 hits 24,485 tok/s/chip, MI300X hits 713. Per-million costs land at $0.026 and $0.37 respectively. GB300 NVL72 is 1313% cheaper per token; GB300 NVL72 delivers 3336% more tok/s/chip.
GB300 NVL72 / MI300X on Qwen 3.5 397B-A17B at 437 tok/s/user: 13,807 / 713 tok/s/chip, $0.046 / $0.37 per million tokens. GB300 NVL72 is 697% cheaper per token; GB300 NVL72 delivers 1837% more tok/s/chip. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB300 NVL7257,477.6MI300X712.6 | GB300 NVL7224,484.6MI300X712.6 | GB300 NVL7213,806.7MI300X712.6 |
| Cost ($/M tok) | GB300 NVL72$0.011MI300X$0.370 | GB300 NVL72$0.026MI300X$0.370 | GB300 NVL72$0.046MI300X$0.370 |
| tok/s/MW | GB300 NVL7227,112,059MI300X512,683 | GB300 NVL7211,549,340MI300X512,683 | GB300 NVL726,512,618MI300X512,683 |
| Concurrency | GB300 NVL72~358MI300X~4 | GB300 NVL72~111MI300X~4 | GB300 NVL72~53MI300X~4 |