MiniMax M3 428B — GB200 NVL72 vs GB300 NVL72
Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) on MiniMax M3 428B. 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.
GB200 NVL72 posts 1598 tok/s/GPU for $0.38 per million tokens at 52 tok/s/user on MiniMax M3 428B; GB300 NVL72 posts 2095 tok/s/GPU for $0.34. GB300 NVL72 is 14% cheaper per token; GB300 NVL72 delivers 31% more tok/s/GPU.
Throughput at 87 tok/s/user on MiniMax M3 428B: GB200 NVL72 hits 638 tok/s/GPU, GB300 NVL72 hits 692. Per-million costs land at $1.00 and $1.10 respectively. GB200 NVL72 is 11% cheaper per token; GB300 NVL72 delivers 8% more tok/s/GPU.
GB200 NVL72 / GB300 NVL72 on MiniMax M3 428B at 122 tok/s/user: 211 / 257 tok/s/GPU, $2.86 / $2.82 per million tokens. GB300 NVL72 is 2% cheaper per token; GB300 NVL72 delivers 22% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | GB200 NVL72:1598.2GB300 NVL72:2094.9 | GB200 NVL72:638.3GB300 NVL72:692.4 | GB200 NVL72:210.9GB300 NVL72:257.4 |
| Cost ($/M tok) | GB200 NVL72:$0.382GB300 NVL72:$0.337 | GB200 NVL72:$0.996GB300 NVL72:$1.103 | GB200 NVL72:$2.862GB300 NVL72:$2.816 |
| tok/s/MW | GB200 NVL72:761024GB300 NVL72:997567 | GB200 NVL72:303973GB300 NVL72:329731 | GB200 NVL72:100447GB300 NVL72:122590 |
| Concurrency | GB200 NVL72:~180GB300 NVL72:~452 | GB200 NVL72:~74GB300 NVL72:~78 | GB200 NVL72:~18GB300 NVL72:~22 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.