DeepSeek R1 — B200 vs MI300X Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus MI300X (AMD CDNA 3) 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.
Near the low end of the 25–68 tok/s/user interactivity band — at 36 tok/s/user — B200 runs $0.09 per million tokens on DeepSeek R1 while MI300X runs $0.99. B200 is the cheaper choice by 976%.
On DeepSeek R1 at 47 tok/s/user, the per-million math comes out to $0.12 for B200 and $1.41 for MI300X; B200 delivers 1112% more output per dollar.
At 58 tok/s/user on DeepSeek R1, B200 costs $0.17 per million tokens; MI300X costs $2.32. B200 is 1279% more cost-efficient at this operating point. (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.)
Chip pricing (owning hyperscaler): B200 $1.73/chip/hr · MI300X $0.95/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B200:$0.092MI300X:$0.986 | B200:$0.116MI300X:$1.410 | B200:$0.168MI300X:$2.322 |
| Concurrency | B200:~1947MI300X:~31 | B200:~1357MI300X:~17 | B200:~1050MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.