Kimi K2.5/K2.6 1T — B200 vs B300 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus B300 (NVIDIA Blackwell) on Kimi K2.5/K2.6 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.
Near the low end of the 37–115 tok/s/user interactivity band — at 56 tok/s/user — B200 runs $1.52 per million tokens on Kimi K2.5/K2.6 1T while B300 runs $1.54. B200 is the cheaper choice by 2%.
On Kimi K2.5/K2.6 1T at 76 tok/s/user, the per-million math comes out to $2.34 for B200 and $2.23 for B300; B300 delivers 5% more output per dollar.
At 96 tok/s/user on Kimi K2.5/K2.6 1T, B200 costs $3.41 per million tokens; B300 costs $3.25. B300 is 5% more cost-efficient at this operating point. (Numbers reflect the default 1k/1k · int4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
GPU pricing (owning hyperscaler): B200 $1.95/GPU/hr · B300 $2.34/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 | B200:$1.515B300:$1.538 | B200:$2.344B300:$2.231 | B200:$3.408B300:$3.247 |
| Concurrency | B200:~26B300:~32 | B200:~12B300:~8 | B200:~7B300:~4 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.