| Route | p50 latency | p95 latency | Throughput | Success rate | ShareShare of traffic |
|---|---|---|---|---|---|
| OpenAI direct | 3.04 s | 13.64 s | 134 tok/s | 99.9% | 73% |
| InferHub | 10.11 s | 82.59 s | 54 tok/s | 82.2% | 22% |
| Venice AI direct | 5.55 s | 18.42 s | 99 tok/s | 99.8% | 4% |
| BlazeAPI | 18.42 s | 45.32 s | 40 tok/s | 83.4% | 1% |
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="gpt-6-astra", messages=[{"role": "user", "content": "Hello!"}] )
SURPLUS_BASE_URL
https://api.surplusintelligence.ai/v1