| Route | p50 latency | p95 latency | Throughput | Success rate | ShareShare of traffic |
|---|---|---|---|---|---|
| BytePlus ModelArk | 5.55 s | 82.59 s | 54 tok/s | 100.0% | 59% |
| Alibaba DashScope — China | 24.87 s | 111.50 s | 30 tok/s | 99.9% | 34% |
| Ollama Cloud direct | 13.64 s | 45.32 s | 73 tok/s | 100.0% | 4% |
| Venice AI direct | 13.64 s | 111.50 s | 40 tok/s | 100.0% | 2% |
| OpenRouter | 13.64 s | 61.18 s | 99 tok/s | 100.0% | 1% |
| Mordiem | 33.57 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="deepseek-v4-pro-0813", messages=[{"role": "user", "content": "Hello!"}] )
SURPLUS_BASE_URL
https://api.surplusintelligence.ai/v1