DeepSeek V4 Pro 1.6T — B200 vs MI355X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on DeepSeek V4 Pro 1.6T. 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.
Throughput at 34 tok/s/user on DeepSeek V4 Pro 1.6T: B200 hits 2880 tok/s/GPU, MI355X hits 911. Per-million costs land at $0.19 and $0.45 respectively. B200 is 139% cheaper per token; B200 delivers 216% more tok/s/GPU.
B200 / MI355X on DeepSeek V4 Pro 1.6T at 65 tok/s/user: 1668 / 340 tok/s/GPU, $0.33 / $1.22 per million tokens. B200 is 276% cheaper per token; B200 delivers 391% more tok/s/GPU.
Toward the upper edge of the 4–127 tok/s/user interactivity band, at 97 tok/s/user on DeepSeek V4 Pro 1.6T: B200 runs 210 tok/s/GPU at $2.62/M tokens, MI355X runs 201 at $2.03/M. MI355X is 29% cheaper per token; B200 delivers 5% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:2880.2MI355X:911.4 | B200:1667.8MI355X:339.9 | B200:209.8MI355X:200.5 |
| Cost ($/M tok) | B200:$0.189MI355X:$0.451 | B200:$0.325MI355X:$1.224 | B200:$2.616MI355X:$2.032 |
| tok/s/MW | B200:1327286MI355X:343907 | B200:768580MI355X:128274 | B200:96669MI355X:75667 |
| Concurrency | B200:~466MI355X:~126 | B200:~121MI355X:~23 | B200:~10MI355X:~9 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.