Grok Chat Completions

Call Grok models through HaoAI’s Grok / xAI API family. The Grok / xAI route selects only Grok-family models and upstreams; its text request and response payloads use xAI’s OpenAI-compatible Chat Completions JSON and SSE shape.

This is a Grok / xAI protocol boundary, not an OpenAI-upstream alias. Use your HaoAI API key—not an xAI OAuth credential—and the Grok endpoint below.

Base URL and endpoint

The protocol root is:


https://api.hao.ai/grok

For the official OpenAI SDK, configure this SDK base URL:


https://api.hao.ai/grok/v1

The Chat Completions endpoint is:


POST https://api.hao.ai/grok/v1/chat/completions

Authentication

Pass your HaoAI API key as a Bearer token:


Authorization: Bearer <your HAOAI_API_KEY>

Available models

Use canonical model IDs with the grok/ prefix.

Model ID	Current use
`grok/grok-4.5`	Default frontier model for general, coding, and agentic work; 500,000-token context window.
`grok/grok-composer-2.5-fast`	Text-only coding model for Grok Build workflows; 200,000-token context window. It does not accept image input.

Request examples

cURL

Terminal


curl https://api.hao.ai/grok/v1/chat/completions \
  -H "Authorization: Bearer $HAOAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok/grok-4.5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Explain how an API gateway works."}
    ]
  }'

Python

grok_chat.py


from openai import OpenAI
 
client = OpenAI(
    base_url="https://api.hao.ai/grok/v1",
    api_key="<your HAOAI_API_KEY>",
)
 
response = client.chat.completions.create(
    model="grok/grok-4.5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain how an API gateway works."},
    ],
)
 
print(response.choices[0].message.content)

TypeScript

grok-chat.ts


import OpenAI from 'openai'
 
const client = new OpenAI({
  baseURL: 'https://api.hao.ai/grok/v1',
  apiKey: '<your HAOAI_API_KEY>'
})
 
const response = await client.chat.completions.create({
  model: 'grok/grok-4.5',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Explain how an API gateway works.' }
  ]
})
 
console.log(response.choices[0].message.content)

Streaming

Set stream: true to receive an SSE stream in the xAI-compatible Chat Completions format:

grok_stream.py


stream = client.chat.completions.create(
    model="grok/grok-4.5",
    messages=[{"role": "user", "content": "Write a short welcome message."}],
    stream=True,
)
 
for chunk in stream:
    content = chunk.choices[0].delta.content
    if content:
        print(content, end="", flush=True)

For the xAI-compatible Responses shape, use the Grok Responses API.