Quickstart

This is the fastest path from zero to a working text chat call. Everything below uses the currently callable API.

1. Get a key

export BEATRA_API_KEY="<YOUR_KEY>"
export BEATRA_BASE_URL="https://api.beatra.ai/v1"

2. Send your first request

curl "$BEATRA_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $BEATRA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "model": "auto",
    "messages": [{"role": "user", "content": "Say hi in five words."}]
  }'

from openai import OpenAI
 
client = OpenAI(
    base_url="https://api.beatra.ai/v1",
    api_key="<YOUR_KEY>",
)
 
resp = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Say hi in five words."}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";
 
const client = new OpenAI({
  baseURL: "https://api.beatra.ai/v1",
  apiKey: process.env.BEATRA_API_KEY,
});
 
const response = await client.chat.completions.create({
  model: "auto",
  messages: [{ role: "user", content: "Say hi in five words." }],
});
console.log(response.choices[0].message.content);

3. What you get back

{
  "id": "chatcmpl_01J5...",
  "model": "<the model that handled the request>",
  "choices": [
    {
      "message": { "role": "assistant", "content": "Hi there my friend!" },
      "finish_reason": "stop"
    }
  ]
}

Keep two things from every response:

The model field — it tells you which model actually handled the request, useful for support and evaluation.
The X-Request-Id response header — store it next to the user-visible job; it's the fastest support lookup key.

If something goes wrong

A bad key returns:

{
  "error": {
    "code": "invalid_api_key",
    "message": "API key not found, signature mismatch, disabled, or revoked.",
    "retryable": false
  }
}

retryable: false means don't loop — fix the request before trying again. The full rule lives in Errors & retries.

Music generation, single-shot uploads, and task lookup are callable Preview APIs. Start from Music generation or the API Reference. Image, video, speech, voices, resumable uploads, and customer callbacks are still Planned.