DeepSeek V4 Pro 1.6T — B300 vs MI355X
Head-to-head AI inference benchmark comparison of B300 (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.
B300 posts 1592 tok/s/GPU for $0.42 per million tokens at 36 tok/s/user on DeepSeek V4 Pro 1.6T; MI355X posts 832 tok/s/GPU for $0.50. B300 is 19% cheaper per token; B300 delivers 91% more tok/s/GPU.
Throughput at 67 tok/s/user on DeepSeek V4 Pro 1.6T: B300 hits 437 tok/s/GPU, MI355X hits 320. Per-million costs land at $1.48 and $1.30 respectively. MI355X is 14% cheaper per token; B300 delivers 37% more tok/s/GPU.
B300 / MI355X on DeepSeek V4 Pro 1.6T at 97 tok/s/user: 254 / 201 tok/s/GPU, $2.59 / $2.03 per million tokens. MI355X is 27% cheaper per token; B300 delivers 27% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B300:1591.5MI355X:832.0 | B300:437.0MI355X:319.7 | B300:254.1MI355X:200.5 |
| Cost ($/M tok) | B300:$0.417MI355X:$0.498 | B300:$1.478MI355X:$1.301 | B300:$2.590MI355X:$2.032 |
| tok/s/MW | B300:733422MI355X:313961 | B300:201394MI355X:120649 | B300:117091MI355X:75667 |
| Concurrency | B300:~90MI355X:~107 | B300:~13MI355X:~21 | B300:~6MI355X:~9 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.