DeepSeek R1 — B200 vs B300 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus B300 (NVIDIA Blackwell) on DeepSeek R1. Owning-hyperscaler TCO normalized by output tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
Push DeepSeek R1 to 72 tok/s/user and B200 lands at $0.19 per million tokens against B300's $0.17 — B300 pulls ahead by 16%.
B200: $0.54 per million tokens. B300: $0.77. Both at 135 tok/s/user on DeepSeek R1, with B200 44% cheaper.
Toward the upper edge of the 9–260 tok/s/user interactivity band — at 198 tok/s/user — B200 runs $1.52 per million tokens on DeepSeek R1 while B300 runs $1.00. B300 is the cheaper choice by 53%. (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.)
GPU pricing (owning hyperscaler): B200 $1.95/GPU/hr · B300 $2.34/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B200:$0.194B300:$0.168 | B200:$0.537B300:$0.772 | B200:$1.522B300:$0.996 |
| Concurrency | B200:~208B300:~95 | B200:~13B300:~10 | B200:~2B300:~18 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.