DeepSeek R1 — B300 vs MI300X Performance per Dollar
Cost per million tokens of B300 (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.
On DeepSeek R1 at 36 tok/s/user, the per-million math comes out to $0.16 for B300 and $1.16 for MI300X; B300 delivers 630% more output per dollar.
At 47 tok/s/user on DeepSeek R1, B300 costs $0.18 per million tokens; MI300X costs $1.66. B300 is 819% more cost-efficient at this operating point.
B300 edges MI300X at 58 tok/s/user on DeepSeek R1 — $0.30 per million tokens versus $2.74, a 817% cost-per-token gap. (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.)
GPU pricing (owning hyperscaler): B300 $2.34/GPU/hr · MI300X $1.12/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 | B300:$0.159MI300X:$1.163 | B300:$0.181MI300X:$1.663 | B300:$0.299MI300X:$2.737 |
| Concurrency | B300:~2795MI300X:~31 | B300:~2073MI300X:~17 | B300:~1094MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.