Qwen 3.5 397B-A17B · Chip comparison
Qwen 3.5 397B-A17B — H200 vs MI300X
Head-to-head AI inference benchmark comparison of H200 (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.
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 →
Throughput at 78 tok/s/user on Qwen 3.5 397B-A17B: H200 hits 19,614 tok/s/chip, MI300X hits 9,164. Costs per million tokens land at $0.017 and $0.029 respectively. H200 is 40% cheaper per token; H200 delivers 114% more tok/s/chip.
H200 / MI300X on Qwen 3.5 397B-A17B at 118 tok/s/user: 17,602 / 5,897 tok/s/chip, $0.019 / $0.045 per million tokens. H200 is 57% cheaper per token; H200 delivers 198% more tok/s/chip.
Toward the upper edge of the 39–198 tok/s/user interactivity band, at 158 tok/s/user on Qwen 3.5 397B-A17B: H200 runs 14,590 tok/s/chip at $0.023 per million tokens, MI300X runs 3,915 at $0.067. H200 is 66% cheaper per token; H200 delivers 3.7× the tok/s/chip. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H20019,614MI300X9,164 | H20017,602MI300X5,897 | H20014,590MI300X3,915 |
| Cost ($/M tok) | H200$0.017MI300X$0.029 | H200$0.019MI300X$0.045 | H200$0.023MI300X$0.067 |
| tok/s/MW | H20014,316,581MI300X6,593,123 | H20012,848,483MI300X4,242,607 | H20010,649,328MI300X2,816,539 |
| Concurrency | H200~34MI300X~20 | H200~26MI300X~17 | H200~19MI300X~11 |