Moonshot AIAPI Host Benchmarks
Kimi K2 Thinking — API Providers & Speed Matrix
Compare live throughput, TTFT latency distributions, prompt cache pricing, and hardware stacks across cloud hosting endpoints.
Fastest Provider
500 t/s
Lowest Latency (TTFT)
0.09s P50
Best Price Value
$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 | 0.8s (1.12s p90) | $1 | $4 | $0.1 | $1.75 | 99.99% |
Groq Cloud | Groq LPU v2 (SRAM) | 168 t/s | 0.09s (0.12s p90) | $0.85 | $3.4 | $0.08 | $1.49 | 99.95% |
Cerebras Cloud | CS-3 Wafer-Scale Engine | 500 t/s | 0.12s (0.16s p90) | $0.9 | $3.6 | — | $1.58 | 99.90% |
Together AI | Nvidia H100 SXM5 | 60 t/s | 0.64s (0.88s p90) | $0.65 | $2.8 | $0.06 | $1.19 | 99.95% |
DeepInfra | Nvidia H100 / L40S | 52 t/s | 0.72s (1s p90) | $0.45 | $2 | $0.05 | $0.84 | 99.90% |
Fireworks AI | Nvidia B200 / H100 | 76 t/s | 0.56s (0.76s p90) | $0.6 | $2.6 | $0.06 | $1.1 | 99.99% |