DeepSeek R1 — GB300 NVL72 vs MI355X Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) 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 66 tok/s/user, the per-million math comes out to $0.18 for GB300 NVL72 and $0.50 for MI355X; GB300 NVL72 delivers 177% more output per dollar.
At 116 tok/s/user on DeepSeek R1, GB300 NVL72 costs $0.48 per million tokens; MI355X costs $1.43. GB300 NVL72 is 194% more cost-efficient at this operating point.
MI355X edges GB300 NVL72 at 166 tok/s/user on DeepSeek R1 — $3.64 per million tokens versus $3.73, a 3% 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.)
Chip pricing (owning hyperscaler): GB300 NVL72 $2.31/chip/hr · MI355X $1.50/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 | GB300 NVL72:$0.181MI355X:$0.503 | GB300 NVL72:$0.484MI355X:$1.426 | GB300 NVL72:$3.734MI355X:$3.637 |
| Concurrency | GB300 NVL72:~1151MI355X:~115 | GB300 NVL72:~337MI355X:~10 | GB300 NVL72:~32MI355X:~4 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.