Quickstart

Router Roo is a drop-in replacement for the OpenAI API. Point any OpenAI-compatible client at our base URL, use your Router Roo key, and every model in the catalog is available under one bill.

1. Create an API key

2. Send your first request

curl

curl https://api.rooter.dev/v1/chat/completions \
  -H "Authorization: Bearer sk-rooter-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "rooter-smart-v1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

JavaScript — openai SDK

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.rooter.dev/v1",
  apiKey: process.env.ROOTER_API_KEY,
});

const res = await client.chat.completions.create({
  model: "rooter-smart-v1",
  messages: [{ role: "user", content: "Hello!" }],
});

Python — openai SDK

from openai import OpenAI

client = OpenAI(
    base_url="https://api.rooter.dev/v1",
    api_key=os.environ["ROOTER_API_KEY"],
)

res = client.chat.completions.create(
    model="rooter-smart-v1",
    messages=[{"role": "user", "content": "Hello!"}],
)

3. Choose a model

Use rooter-smart-v1 (the Smart Router) to have every request routed to the best model for the job, or pin any catalog model by id — see benchmarks for how they compare. Pricing per model is on the pricing page.

Guard rails

Set daily spend limits per key or account-wide under Limits, enable auto top-up so long-running jobs never stall, and cap teammate spending with per-member limits under Settings → Team.

Try it without writing code

The dashboard Playground lets you test every model straight from your browser, billed to your credit — no key required.

Open Playground