DeepSeek V4 Pro 1.6T — B200 vs B300 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus B300 (NVIDIA Blackwell) on DeepSeek V4 Pro 1.6T. 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.
B200: $0.41 per million tokens. B300: $0.77. Both at 44 tok/s/user on DeepSeek V4 Pro 1.6T, with B200 88% cheaper.
Around the middle of the 5–161 tok/s/user interactivity band — at 83 tok/s/user — B200 runs $2.35 per million tokens on DeepSeek V4 Pro 1.6T while B300 runs $1.71. B300 is the cheaper choice by 37%.
On DeepSeek V4 Pro 1.6T at 122 tok/s/user, the per-million math comes out to $5.43 for B200 and $4.28 for B300; B300 delivers 27% more output per dollar. (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.)
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:$0.413B300:$0.774 | B200:$2.348B300:$1.714 | B200:$5.429B300:$4.279 |
| Concurrency | B200:~137B300:~41 | B200:~12B300:~10 | B200:~4B300:~2 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.