DeepSeek R1 — B200 vs GB300 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) on DeepSeek R1. 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 70 tok/s/user as the target on DeepSeek R1, B200 produces 1058 tok/s/chip ($0.46 per million tokens) and GB300 NVL72 produces 3521 ($0.18). GB300 NVL72 is 151% cheaper per token; GB300 NVL72 delivers 233% more tok/s/chip.
At 127 tok/s/user interactivity on DeepSeek R1, B200 delivers 275 tok/s/chip at $1.75 per million tokens; GB300 NVL72 delivers 848 tok/s/chip at $0.75. GB300 NVL72 is 133% cheaper per token; GB300 NVL72 delivers 208% more tok/s/chip at this point.
B200 posts 157 tok/s/chip for $3.04 per million tokens at 184 tok/s/user on DeepSeek R1; GB300 NVL72 posts 117 tok/s/chip for $5.58. B200 is 83% cheaper per token; B200 delivers 34% more tok/s/chip. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:1058.4GB300 NVL72:3521.4 | B200:274.8GB300 NVL72:847.5 | B200:156.8GB300 NVL72:117.0 |
| Cost ($/M tok) | B200:$0.458GB300 NVL72:$0.183 | B200:$1.751GB300 NVL72:$0.750 | B200:$3.042GB300 NVL72:$5.578 |
| tok/s/MW | B200:618928GB300 NVL72:1661035 | B200:160719GB300 NVL72:399769 | B200:91701GB300 NVL72:55169 |
| Concurrency | B200:~62GB300 NVL72:~977 | B200:~8GB300 NVL72:~188 | B200:~25GB300 NVL72:~20 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.