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.
Setting 81 tok/s/user as the target on Qwen 3.5 397B-A17B, B300 produces 2395 tok/s/GPU ($0.27 per million tokens) and MI355X produces 1625 ($0.26). MI355X is 6% cheaper per token; B300 delivers 47% more tok/s/GPU.
At 133 tok/s/user interactivity on Qwen 3.5 397B-A17B, B300 delivers 1152 tok/s/GPU at $0.56 per million tokens; MI355X delivers 913 tok/s/GPU at $0.45. MI355X is 25% cheaper per token; B300 delivers 26% more tok/s/GPU at this point.
B300 posts 752 tok/s/GPU for $0.86 per million tokens at 185 tok/s/user on Qwen 3.5 397B-A17B; MI355X posts 510 tok/s/GPU for $0.78. MI355X is 11% cheaper per token; B300 delivers 47% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B300:2394.8MI355X:1625.2 | B300:1152.2MI355X:913.0 | B300:751.9MI355X:510.4 |
| Cost ($/M tok) | B300:$0.272MI355X:$0.257 | B300:$0.564MI355X:$0.453 | B300:$0.865MI355X:$0.781 |
| tok/s/MW | B300:1103604MI355X:613278 | B300:530958MI355X:344529 | B300:346508MI355X:192596 |
| Concurrency | B300:~61MI355X:~41 | B300:~18MI355X:~14 | B300:~9MI355X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.