Kimi K2.5/K2.6/K2.7-Code 1T — B200 vs GB300 NVL72 Performance per Dollar
Cost per million tokens of B200 (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.
On Kimi K2.5/K2.6/K2.7-Code 1T at 63 tok/s/user, the per-million math comes out to $0.13 for B200 and $0.06 for GB300 NVL72; GB300 NVL72 delivers 102% more output per dollar.
At 103 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T, B200 costs $0.27 per million tokens; GB300 NVL72 costs $0.38. B200 is 41% more cost-efficient at this operating point.
B200 edges GB300 NVL72 at 143 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T — $0.82 per million tokens versus $0.87, a 6% cost-per-token gap. (Numbers reflect the default 8k/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): B200 $1.73/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 | B200:$0.127GB300 NVL72:$0.063 | B200:$0.270GB300 NVL72:$0.382 | B200:$0.822GB300 NVL72:$0.869 |
| Concurrency | B200:~231GB300 NVL72:~1202 | B200:~77GB300 NVL72:~68 | B200:~10GB300 NVL72:~26 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.