DeepSeek V4 Pro 1.6T · GPU comparison

DeepSeek V4 Pro 1.6T — MI300X vs MI325X

Head-to-head AI inference benchmark comparison of MI300X (AMD CDNA 3) and MI325X (AMD CDNA 3) 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 12 tok/s/user on DeepSeek V4 Pro 1.6T: MI300X hits 292 tok/s/GPU, MI325X hits 387. Per-million costs land at $1.07 and $0.92 respectively. MI325X is 16% cheaper per token; MI325X delivers 32% more tok/s/GPU.

MI300X / MI325X on DeepSeek V4 Pro 1.6T at 19 tok/s/user: 130 / 207 tok/s/GPU, $2.36 / $1.77 per million tokens. MI325X is 33% cheaper per token; MI325X delivers 59% more tok/s/GPU.

Toward the upper edge of the 5–32 tok/s/user interactivity band, at 25 tok/s/user on DeepSeek V4 Pro 1.6T: MI300X runs 88 tok/s/GPU at $3.54/M tokens, MI325X runs 111 at $3.12/M. MI325X is 13% cheaper per token; MI325X delivers 26% 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.)

View performance-per-dollar view →

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/gpu)
MI300X:292.2MI325X:386.8
MI300X:130.0MI325X:207.3
MI300X:88.2MI325X:111.4
Cost ($/M tok)
MI300X:$1.069MI325X:$0.920
MI300X:$2.357MI325X:$1.771
MI300X:$3.540MI325X:$3.125
tok/s/MW
MI300X:210222MI325X:228896
MI300X:93512MI325X:122634
MI300X:63442MI325X:65918
Concurrency
MI300X:~101MI325X:~136
MI300X:~29MI325X:~46
MI300X:~15MI325X:~18

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.

Vendor:
Aggregation:
Spec Decoding: