Kimi K2.5/K2.6 1T · Performance per Dollar

Kimi K2.5/K2.6 1T — B200 vs GB200 NVL72 Performance per Dollar

Cost per million tokens of B200 (NVIDIA Blackwell) versus GB200 NVL72 (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.

GB200 NVL72 edges B200 at 61 tok/s/user on Kimi K2.5/K2.6 1T — $0.34 per million tokens versus $0.61, a 81% cost-per-token gap.

Push Kimi K2.5/K2.6 1T to 88 tok/s/user and B200 lands at $0.99 per million tokens against GB200 NVL72's $0.79 — GB200 NVL72 pulls ahead by 26%.

B200: $1.74 per million tokens. GB200 NVL72: $2.57. Both at 115 tok/s/user on Kimi K2.5/K2.6 1T, with B200 47% cheaper. (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.)

GPU pricing (owning hyperscaler): B200 $1.95/GPU/hr · GB200 NVL72 $2.21/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.

View full latency + throughput comparison →

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)
Dollar per Million Tokens
B200:$0.612GB200 NVL72:$0.339
B200:$0.993GB200 NVL72:$0.788
B200:$1.742GB200 NVL72:$2.568
Concurrency
B200:~29GB200 NVL72:~667
B200:~12GB200 NVL72:~173
B200:~5GB200 NVL72:~26

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.