gpt-oss 120B — MI300X vs MI325X
Head-to-head AI inference benchmark comparison of MI300X (AMD CDNA 3) and MI325X (AMD CDNA 3) on gpt-oss 120B. 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 50 tok/s/user on gpt-oss 120B: MI300X hits 3834 tok/s/chip, MI325X hits 2737. Per-million costs land at $0.07 and $0.11 respectively. MI300X is 62% cheaper per token; MI300X delivers 40% more tok/s/chip.
MI300X / MI325X on gpt-oss 120B at 71 tok/s/user: 3127 / 2084 tok/s/chip, $0.09 / $0.15 per million tokens. MI300X is 69% cheaper per token; MI300X delivers 50% more tok/s/chip.
Toward the upper edge of the 30–113 tok/s/user interactivity band, at 92 tok/s/user on gpt-oss 120B: MI300X runs 2051 tok/s/chip at $0.13/M tokens, MI325X runs 1076 at $0.28/M. MI300X is 127% cheaper per token; MI300X delivers 91% more tok/s/chip. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI300X:3834.3MI325X:2737.2 | MI300X:3127.3MI325X:2084.1 | MI300X:2050.9MI325X:1075.9 |
| Cost ($/M tok) | MI300X:$0.069MI325X:$0.111 | MI300X:$0.087MI325X:$0.146 | MI300X:$0.126MI325X:$0.285 |
| tok/s/MW | MI300X:2758480MI325X:1619651 | MI300X:2249882MI325X:1233185 | MI300X:1475490MI325X:636617 |
| Concurrency | MI300X:~64MI325X:~64 | MI300X:~49MI325X:~63 | MI300X:~23MI325X:~26 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.