gpt-oss 120B — B200 vs MI300X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI300X (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.
At 106 tok/s/user interactivity on gpt-oss 120B, B200 delivers 20630 tok/s/GPU at $0.03 per million tokens; MI300X delivers 1744 tok/s/GPU at $0.18. B200 is 572% cheaper per token; B200 delivers 1083% more tok/s/GPU at this point.
B200 posts 18370 tok/s/GPU for $0.03 per million tokens at 154 tok/s/user on gpt-oss 120B; MI300X posts 842 tok/s/GPU for $0.36. B200 is 1106% cheaper per token; B200 delivers 2081% more tok/s/GPU.
Throughput at 201 tok/s/user on gpt-oss 120B: B200 hits 9759 tok/s/GPU, MI300X hits 323. Per-million costs land at $0.06 and $0.94 respectively. B200 is 1609% cheaper per token; B200 delivers 2920% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:20630.5MI300X:1743.9 | B200:18370.4MI300X:842.2 | B200:9759.0MI300X:323.1 |
| Cost ($/M tok) | B200:$0.026MI300X:$0.178 | B200:$0.029MI300X:$0.355 | B200:$0.055MI300X:$0.941 |
| tok/s/MW | B200:9507127MI300X:974257 | B200:8465642MI300X:470523 | B200:4497220MI300X:180505 |
| Concurrency | B200:~241MI300X:~17 | B200:~126MI300X:~5 | B200:~64MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.