Kimi K2.5/K2.6/K2.7-Code 1T — B300 vs GB300 NVL72 Performance per Dollar
Cost per million tokens of B300 (NVIDIA Blackwell) versus GB300 NVL72 (NVIDIA Blackwell) on Kimi K2.5/K2.6/K2.7-Code 1T. 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.
At 69 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T, B300 costs $0.67 per million tokens; GB300 NVL72 costs $0.11. GB300 NVL72 is 508% more cost-efficient at this operating point.
B300 edges GB300 NVL72 at 104 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T — $0.70 per million tokens versus $1.12, a 60% cost-per-token gap.
Push Kimi K2.5/K2.6/K2.7-Code 1T to 138 tok/s/user and B300 lands at $1.28 per million tokens against GB300 NVL72's $2.88 — B300 pulls ahead by 125%. (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.)
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.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B300:$0.675GB300 NVL72:$0.111 | B300:$0.699GB300 NVL72:$1.119 | B300:$1.278GB300 NVL72:$2.879 |
| Concurrency | B300:~28GB300 NVL72:~1977 | B300:~19GB300 NVL72:~119 | B300:~7GB300 NVL72:~30 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.