Grok API

Two Grok models are available through the same passthrough as the rest of the Text API, on all four request formats.

Both reason by default and report the thinking budget under usage.completion_tokens_details.reasoning_tokens, which is included in completion_tokens and billed as output.

OpenAI Chat Completions

POST https://api.beatapi.io/v1/chat/completions

1
import OpenAI from "openai";
2
3
const client = new OpenAI({
4
  apiKey: process.env.BEATAPI_API_KEY,
5
  baseURL: "https://api.beatapi.io/v1"
6
});
7
8
const completion = await client.chat.completions.create({
9
  model: "grok-4.6",
10
  max_tokens: 2000,
11
  messages: [
12
    { role: "user", content: "Design a resilient webhook retry strategy for a payments API." }
13
  ]
14
});
15
16
console.log(completion.choices[0].message.content);

$
curl --request POST \
>
  --url https://api.beatapi.io/v1/chat/completions \
>
  --header 'Authorization: Bearer <BEATAPI_API_KEY>' \
>
  --header 'Content-Type: application/json' \
>
  --data '{
>
  "model": "grok-4.6",
>
  "messages": [
>
    {
>
      "role": "user",
>
      "content": "Design a resilient webhook retry strategy for a payments API."
>
    }
>
  ]
>
}'

Long context pricing

Grok switches to a higher rate once the prompt reaches 200,000 tokens, and the higher rate then applies to every token in that request — input, cached input, and output all double. Below the threshold the standard rate applies to the whole request.

Other SDK formats

OpenAI Responses

POST https://api.beatapi.io/v1/responses

$
curl --request POST \
>
  --url https://api.beatapi.io/v1/responses \
>
  --header 'Authorization: Bearer <BEATAPI_API_KEY>' \
>
  --header 'Content-Type: application/json' \
>
  --data '{
>
  "model": "grok-4.6",
>
  "input": "Design a resilient webhook retry strategy for a payments API.",
>
  "reasoning": {
>
    "effort": "medium"
>
  }
>
}'

Anthropic Messages

POST https://api.beatapi.io/v1/messages

$
curl --request POST \
>
  --url https://api.beatapi.io/v1/messages \
>
  --header 'Authorization: Bearer <BEATAPI_API_KEY>' \
>
  --header 'Content-Type: application/json' \
>
  --data '{
>
  "model": "grok-4.6",
>
  "max_tokens": 1024,
>
  "messages": [
>
    {
>
      "role": "user",
>
      "content": "Design a resilient webhook retry strategy for a payments API."
>
    }
>
  ]
>
}'

Gemini-compatible content

POST https://api.beatapi.io/v1beta/models/{model}:generateContent

$
curl --request POST \
>
  --url 'https://api.beatapi.io/v1beta/models/grok-4.6:generateContent' \
>
  --header 'Authorization: Bearer <BEATAPI_API_KEY>' \
>
  --header 'Content-Type: application/json' \
>
  --data '{
>
  "contents": [
>
    {
>
      "role": "user",
>
      "parts": [
>
        {
>
          "text": "Design a resilient webhook retry strategy for a payments API."
>
        }
>
      ]
>
    }
>
  ]
>
}'

Authentication and availability

List the currently enabled text models before rollout:

$
curl https://api.beatapi.io/v1/models \
>
  -H "Authorization: Bearer $BEATAPI_API_KEY"

1
{
2
  "object": "list",
3
  "data": [
4
    { "id": "grok-4.6", "object": "model", "owned_by": "beatapi" }
5
  ]
6
}

The same API key and USD balance are shared across Text, Image, Video, Workflow, Effect, and Realtime APIs. Calls are metered from actual token usage; inspect request IDs, model, token totals, status, and settled amount in Dashboard usage logs.

Production checklist

• Budget output tokens for reasoning, not just for the answer.
• Watch the 200K prompt threshold — crossing it doubles the whole request.
• Keep API keys on trusted servers and redact them from logs.
• Log the response request ID for support and billing audits.
• Handle 401, 402, 429, 502, and 503 as distinct operational cases.

Continue with media models

The same BeatAPI account can call GPT-5.6, Claude, Gemini, and the Image and Video APIs without adding another billing or authentication system.