DeepSeek R1 — MI325X vs Vera Rubin NVL72
Head-to-head AI inference benchmark comparison of MI325X (AMD CDNA 3) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on DeepSeek R1. 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.
Vera Rubin NVL72: 0 tok/s/chip, $0.00 per million tokens at 139 tok/s/user on DeepSeek R1. MI325X is unmeasured here.
At 226 tok/s/user on DeepSeek R1, Vera Rubin NVL72 delivers 0 tok/s/chip at $0.00 per million tokens; MI325X hasn't been benchmarked at this target.
Vera Rubin NVL72 hits 0 tok/s/chip for $0.00 per million tokens at 313 tok/s/user on DeepSeek R1. No MI325X data at this operating point. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI325X:—Vera Rubin NVL72:0.0 | MI325X:—Vera Rubin NVL72:0.0 | MI325X:—Vera Rubin NVL72:0.0 |
| Cost ($/M tok) | MI325X:—Vera Rubin NVL72:$0.000 | MI325X:—Vera Rubin NVL72:$0.000 | MI325X:—Vera Rubin NVL72:$0.000 |
| tok/s/MW | MI325X:—Vera Rubin NVL72:0 | MI325X:—Vera Rubin NVL72:0 | MI325X:—Vera Rubin NVL72:0 |
| Concurrency | MI325X:—Vera Rubin NVL72:~400 | MI325X:—Vera Rubin NVL72:~400 | MI325X:—Vera Rubin NVL72:~400 |