MiniMax M3 428B — B200 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 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.
B200 hits 5518 tok/s/chip for $0.09 per million tokens at 148 tok/s/user on MiniMax M3 428B. No GB200 NVL72 data at this operating point.
B200: 2213 tok/s/chip, $0.22 per million tokens at 290 tok/s/user on MiniMax M3 428B. GB200 NVL72 is unmeasured here.
At 432 tok/s/user on MiniMax M3 428B, B200 delivers 831 tok/s/chip at $0.58 per million tokens; GB200 NVL72 hasn't been benchmarked at this target. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:5518.1GB200 NVL72:— | B200:2212.6GB200 NVL72:— | B200:831.3GB200 NVL72:— |
| Cost ($/M tok) | B200:$0.087GB200 NVL72:— | B200:$0.217GB200 NVL72:— | B200:$0.578GB200 NVL72:— |
| tok/s/MW | B200:3226982GB200 NVL72:— | B200:1293940GB200 NVL72:— | B200:486167GB200 NVL72:— |
| Concurrency | B200:~40GB200 NVL72:— | B200:~4GB200 NVL72:— | B200:~2GB200 NVL72:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.