Moonshot AIAPI Host BenchmarksEstimated
Kimi K2.6 — API Providers & Speed Matrix
Modelled throughput, TTFT latency and prompt-cache pricing estimates across cloud hosting endpoints, derived from official list prices and public benchmarks.
Fastest provider
500 t/s
Lowest latency (TTFT)
90 ms P50
Best price
$0.45 / $2
Blended Pricing Workload Calculator
Switch between standard conversational ratios, agentic caching loops, and heavy RAG workloads.
| Provider & Host | Hardware Stack | Output Speed (TPS) | Latency TTFT (P50) | Input Price ($/1M) | Output Price ($/1M) | Cache Read ($/1M) | Blended Price ($/1M) | SLA |
|---|---|---|---|---|---|---|---|---|
Moonshot AIOfficial | Cloud TPU / Nvidia H100 | 40 t/s | 800 ms (1.12 s p90) | $1 | $4 | $0.1 | $1.75 | 99.99% |
| Groq LPU v2 (SRAM) | 168 t/s | 90 ms (120 ms p90) | $0.85 | $3.4 | $0.08 | $1.49 | 99.95% | |
| CS-3 Wafer-Scale Engine | 500 t/s | 120 ms (160 ms p90) | $0.9 | $3.6 | — | $1.58 | 99.90% | |
| Nvidia H100 SXM5 | 60 t/s | 640 ms (880 ms p90) | $0.65 | $2.8 | $0.06 | $1.19 | 99.95% | |
| Nvidia H100 / L40S | 52 t/s | 720 ms (1.00 s p90) | $0.45 | $2 | $0.05 | $0.84 | 99.90% | |
| Nvidia B200 / H100 | 76 t/s | 560 ms (760 ms p90) | $0.6 | $2.6 | $0.06 | $1.1 | 99.99% |
Figures are modelled estimates based on published list prices and public throughput benchmarks — not measured on live endpoints. Verify current pricing and terms with each host before committing.