Moonshot AIAPI Host BenchmarksEstimated
Kimi K3 — 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
463 t/s
Lowest latency (TTFT)
90 ms P50
Best price
$1.35 / $7.5
Blended Pricing Workload Calculator
Switch between standard conversational ratios, agentic caching loops, and heavy RAG workloads.
| Provider & Host | Hardware Stack | Output Speed (TPS) | Latency TTFT (P50) | Input Price ($/1M) | Output Price ($/1M) | Cache Read ($/1M) | Blended Price ($/1M) | SLA |
|---|---|---|---|---|---|---|---|---|
Moonshot AIOfficial | Cloud TPU / Nvidia H100 | 37 t/s | 3.06 s (4.28 s p90) | $3 | $15 | $0.3 | $6 | 99.99% |
| Groq LPU v2 (SRAM) | 155 t/s | 90 ms (120 ms p90) | $2.55 | $12.75 | $0.24 | $5.1 | 99.95% | |
| CS-3 Wafer-Scale Engine | 463 t/s | 120 ms (160 ms p90) | $2.7 | $13.5 | — | $5.4 | 99.90% | |
| Nvidia H100 SXM5 | 56 t/s | 2.45 s (3.37 s p90) | $1.95 | $10.5 | $0.18 | $4.09 | 99.95% | |
| Nvidia H100 / L40S | 48 t/s | 2.75 s (3.83 s p90) | $1.35 | $7.5 | $0.15 | $2.89 | 99.90% | |
| Nvidia B200 / H100 | 70 t/s | 2.14 s (2.91 s p90) | $1.8 | $9.75 | $0.18 | $3.79 | 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.