MiniMax M2.5/M2.7 — B200 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on MiniMax M2.5/M2.7. 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.
Setting 68 tok/s/user as the target on MiniMax M2.5/M2.7, B200 produces 8180 tok/s/GPU ($0.07 per million tokens) and GB200 NVL72 produces 8433 ($0.07). B200 is 10% cheaper per token; GB200 NVL72 delivers 3% more tok/s/GPU.
At 105 tok/s/user interactivity on MiniMax M2.5/M2.7, B200 delivers 2983 tok/s/GPU at $0.18 per million tokens; GB200 NVL72 delivers 3247 tok/s/GPU at $0.19. B200 is 5% cheaper per token; GB200 NVL72 delivers 9% more tok/s/GPU at this point.
B200 posts 1152 tok/s/GPU for $0.47 per million tokens at 143 tok/s/user on MiniMax M2.5/M2.7; GB200 NVL72 posts 1066 tok/s/GPU for $0.58. B200 is 23% cheaper per token; B200 delivers 8% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:8180.3GB200 NVL72:8432.8 | B200:2983.2GB200 NVL72:3246.9 | B200:1151.7GB200 NVL72:1066.4 |
| Cost ($/M tok) | B200:$0.066GB200 NVL72:$0.073 | B200:$0.182GB200 NVL72:$0.190 | B200:$0.470GB200 NVL72:$0.581 |
| tok/s/MW | B200:3769706GB200 NVL72:4015616 | B200:1374764GB200 NVL72:1546138 | B200:530757GB200 NVL72:507786 |
| Concurrency | B200:~831GB200 NVL72:~928 | B200:~105GB200 NVL72:~236 | B200:~10GB200 NVL72:~45 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.