DeepSeek R1 · Performance per Dollar

DeepSeek R1 — B300 vs GB300 NVL72 Performance per Dollar

Cost per million tokens of B300 (NVIDIA Blackwell) versus GB300 NVL72 (NVIDIA Blackwell) on DeepSeek R1. Large-hyperscaler-volume ownership TCO normalized by total 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 73 tok/s/user, the per-million math comes out to $0.17 for B300 and $0.092 for GB300 NVL72; GB300 NVL72 delivers 82% more total tokens per dollar.

At 126 tok/s/user on DeepSeek R1, B300 costs $0.56 per million tokens; GB300 NVL72 costs $0.17. GB300 NVL72 delivers 3.3× the tokens per dollar at this operating point.

B300 edges GB300 NVL72 at 178 tok/s/user on DeepSeek R1 — $0.79 per million tokens versus $1.18, a 33% lower cost per token. (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.)

Chip pricing (owning hyperscaler): B300 $2.26/chip/hr · GB300 NVL72 $2.31/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.

View full latency and throughput comparison →

GB300 NVL72 vs B300: full specs & benchmarks across models →

Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.

Metric
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Dollars per Million Tokens
B300$0.17GB300 NVL72$0.092
B300$0.56GB300 NVL72$0.17
B300$0.79GB300 NVL72$1.18
Concurrency
B300~88GB300 NVL72~679
B300~10GB300 NVL72~253
B300~4GB300 NVL72~20
DeepSeek R1: B300 versus GB300 NVL72 cost per million tokens at matched interactivity levels
B300 versus GB300 NVL72 cost per million tokens for this comparison's canonical default workload. Lower cost indicates better performance per dollar.
Open full-size image ↗