Kimi K2.5/K2.6/K2.7-Code 1T — B200 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Kimi K2.5/K2.6/K2.7-Code 1T. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
Setting 64 tok/s/user as the target on Kimi K2.5/K2.6/K2.7-Code 1T, B200 produces 3754 tok/s/chip ($0.13 per million tokens) and GB200 NVL72 produces 8980 ($0.06). GB200 NVL72 is 122% cheaper per token; GB200 NVL72 delivers 139% more tok/s/chip.
At 103 tok/s/user interactivity on Kimi K2.5/K2.6/K2.7-Code 1T, B200 delivers 1594 tok/s/chip at $0.30 per million tokens; GB200 NVL72 delivers 1545 tok/s/chip at $0.33. B200 is 11% cheaper per token; B200 delivers 3% more tok/s/chip at this point.
B200 posts 623 tok/s/chip for $0.77 per million tokens at 141 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T; GB200 NVL72 posts 733 tok/s/chip for $0.70. GB200 NVL72 is 9% cheaper per token; GB200 NVL72 delivers 18% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:3753.8GB200 NVL72:8979.7 | B200:1594.1GB200 NVL72:1545.4 | B200:622.7GB200 NVL72:732.9 |
| Cost ($/M tok) | B200:$0.128GB200 NVL72:$0.058 | B200:$0.301GB200 NVL72:$0.334 | B200:$0.772GB200 NVL72:$0.705 |
| tok/s/MW | B200:2195224GB200 NVL72:4801979 | B200:932248GB200 NVL72:826425 | B200:364180GB200 NVL72:391932 |
| Concurrency | B200:~204GB200 NVL72:~1203 | B200:~45GB200 NVL72:~167 | B200:~2GB200 NVL72:~27 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
Total Tokens per $1 USD (Owning - Hyperscaler) vs. Interactivity
Kimi K2.5/2.6/2.7-Code 1T • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
TCO $/chip/hr: H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
No data available
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip