Qwen 3.5 397B-A17B · Chip comparison

Qwen 3.5 397B-A17B — B300 vs MI355X

Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) 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.

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 →

Setting 119 tok/s/user as the target on Qwen 3.5 397B-A17B, B300 produces 46,014 tok/s/chip ($0.014 per million tokens) and MI355X produces 34,356 ($0.012). MI355X is 11% cheaper per token; B300 delivers 34% more tok/s/chip.

At 207 tok/s/user interactivity on Qwen 3.5 397B-A17B, B300 delivers 20,047 tok/s/chip at $0.031 per million tokens; MI355X delivers 16,202 tok/s/chip at $0.026. MI355X is 18% cheaper per token; B300 delivers 24% more tok/s/chip at this point.

B300 posts 9,536 tok/s/chip for $0.066 per million tokens at 295 tok/s/user on Qwen 3.5 397B-A17B; MI355X posts 7,785 tok/s/chip for $0.054. MI355X is 19% cheaper per token; B300 delivers 22% 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.)

View performance per dollar →

MI355X vs B300: full specs & benchmarks across models →

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)
B30046,014MI355X34,356
B30020,047MI355X16,202
B3009,536MI355X7,785
Cost ($/M tok)
B300$0.014MI355X$0.012
B300$0.031MI355X$0.026
B300$0.066MI355X$0.054
tok/s/MW
B30024,217,888MI355X16,438,134
B30010,551,002MI355X7,752,271
B3005,019,192MI355X3,724,840
Concurrency
B300~41MI355X~26
B300~32MI355X~11
B300~12MI355X~5