MiniMax M3 428B — B200 vs GB300 NVL72
Head-to-head AI inference benchmark comparison of B200 (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.
At 148 tok/s/user on MiniMax M3 428B, B200 delivers 5518 tok/s/chip at $0.09 per million tokens; GB300 NVL72 hasn't been benchmarked at this target.
B200 hits 2213 tok/s/chip for $0.22 per million tokens at 290 tok/s/user on MiniMax M3 428B. No GB300 NVL72 data at this operating point.
B200: 831 tok/s/chip, $0.58 per million tokens at 432 tok/s/user on MiniMax M3 428B. GB300 NVL72 is unmeasured here. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:5518.1GB300 NVL72:— | B200:2212.6GB300 NVL72:— | B200:831.3GB300 NVL72:— |
| Cost ($/M tok) | B200:$0.088GB300 NVL72:— | B200:$0.217GB300 NVL72:— | B200:$0.579GB300 NVL72:— |
| tok/s/MW | B200:3226982GB300 NVL72:— | B200:1293940GB300 NVL72:— | B200:486167GB300 NVL72:— |
| Concurrency | B200:~40GB300 NVL72:— | B200:~4GB300 NVL72:— | B200:~2GB300 NVL72:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.