| Route | p50 latency | p95 latency | Throughput | Success rate | ShareShare of traffic |
|---|---|---|---|---|---|
| Alibaba DashScope — China | 7.49 s | 61.18 s | 30 tok/s | 99.6% | 86% |
| InferHub | 10.11 s | 61.18 s | 40 tok/s | 96.9% | 11% |
| Venice AI direct | 1.24 s | 18.42 s | - | 99.7% | 1% |
| Kenari Cloud | 24.87 s | 150.53 s | 54 tok/s | 99.7% | 1% |
| Baseten | 3.04 s | 45.32 s | - | 100.0% | 1% |
| OpenRouter | 1.67 s | - | - | 83.3% | 0% |
| Ollama Cloud direct | 3.04 s | - | - | 100.0% | 0% |
Any OpenAI-compatible client works. Add /min30 to the path to only buy at 30% off or better.
Read the quickstart ↗from openai import OpenAI client = OpenAI( api_key="inf_your_key_here", base_url="$SURPLUS_BASE_URL" ) response = client.chat.completions.create( model="kimi-k3", messages=[{"role": "user", "content": "Hello!"}] )
SURPLUS_BASE_URL
https://api.surplusintelligence.ai/v1