Home / Docs / Quickstart

Quickstart

This guide gets you from zero to your first observed, governed request in about five minutes.

1. Create a project key

From the dashboard, open Settings → API keys and create a key scoped to chat:write. Copy it — it's shown only once.

shell

# keep keys in env, never in source
export APIHUB_KEY="ah_live_•••••••"

2. Point your SDK at the gateway

APIHub Star is OpenAI-compatible. Change only the base_url:

quickstart.py

from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.apihubstar.tech/v1",
    api_key=os.environ["APIHUB_KEY"],
)

resp = client.chat.completions.create(
    model="auto",
    messages=[{"role":"user","content":"Hello, gateway!"}],
)
print(resp.choices[0].message.content)

3. Or call it over plain HTTP

curl

curl https://api.apihubstar.tech/v1/chat/completions \
  -H "Authorization: Bearer $APIHUB_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"auto","messages":[{"role":"user","content":"hi"}]}'

The special model "auto" lets APIHub Star pick the cheapest upstream that meets your latency target. Pin a provider any time by passing its name instead.

4. Watch it live

Open Observability → Live tail to see the request you just made, with its route, cache status, token cost and latency breakdown.

NextAuthentication → BackDocs overview →