DeepSeek R1 — B200 vs B300
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and B300 (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.
B200 posts 2821 tok/s/GPU for $0.19 per million tokens at 72 tok/s/user on DeepSeek R1; B300 posts 3887 tok/s/GPU for $0.17. B300 is 16% cheaper per token; B300 delivers 38% more tok/s/GPU.
Throughput at 135 tok/s/user on DeepSeek R1: B200 hits 1011 tok/s/GPU, B300 hits 851. Per-million costs land at $0.54 and $0.77 respectively. B200 is 44% cheaper per token; B200 delivers 19% more tok/s/GPU.
B200 / B300 on DeepSeek R1 at 198 tok/s/user: 373 / 660 tok/s/GPU, $1.52 / $1.00 per million tokens. B300 is 53% cheaper per token; B300 delivers 77% more tok/s/GPU. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:2820.9B300:3886.8 | B200:1011.4B300:850.7 | B200:373.2B300:660.2 |
| Cost ($/M tok) | B200:$0.194B300:$0.168 | B200:$0.537B300:$0.772 | B200:$1.522B300:$0.996 |
| tok/s/MW | B200:1299963B300:1791138 | B200:466074B300:392025 | B200:171963B300:304230 |
| Concurrency | B200:~208B300:~95 | B200:~13B300:~10 | B200:~2B300:~18 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.