> ## Documentation Index
> Fetch the complete documentation index at: https://docs.parallellabs.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Use Parallel AI as an OpenAI API Drop-In Replacement

> Parallel AI's chat completions endpoint is fully OpenAI-compatible. Point any OpenAI SDK at your Parallel AI endpoint and start using it immediately.

## Overview

Parallel AI's `POST /api/v0/chat/completions` endpoint is 100% compatible with OpenAI's Chat Completions API spec. To migrate an existing integration, you only need to change the `base_url` in your client — no other code changes are required.

<Tip>
  This means any library, tool, or service built for the OpenAI API works out of
  the box with Parallel AI. Swap the base URL and your API key, and you're done.
</Tip>

## Quick Start

<CodeGroup>
  ```python Python theme={null}
  from openai import OpenAI

  client = OpenAI(
      api_key="YOUR_API_KEY",
      base_url="https://api.parallellabs.app/api/v0"
  )

  response = client.chat.completions.create(
      model="gpt-4o",
      messages=[
          {"role": "system", "content": "You are a helpful assistant."},
          {"role": "user", "content": "What can Parallel AI help me with?"}
      ]
  )

  print(response.choices[0].message.content)
  ```

  ```javascript Node.js theme={null}
  import OpenAI from 'openai';

  const client = new OpenAI({
    apiKey: 'YOUR_API_KEY',
    baseURL: 'https://api.parallellabs.app/api/v0',
  });

  const response = await client.chat.completions.create({
    model: 'gpt-4o',
    messages: [
      { role: 'system', content: 'You are a helpful assistant.' },
      { role: 'user', content: 'What can Parallel AI help me with?' },
    ],
  });

  console.log(response.choices[0].message.content);
  ```

  ```bash cURL theme={null}
  curl -X POST "https://api.parallellabs.app/api/v0/chat/completions" \
    -H "X-API-Key: YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "gpt-4o",
      "messages": [
        { "role": "system", "content": "You are a helpful assistant." },
        { "role": "user", "content": "What can Parallel AI help me with?" }
      ]
    }'
  ```
</CodeGroup>

## Supported Parameters

The following request parameters are fully supported:

| Parameter           | Type             | Description                                             |
| ------------------- | ---------------- | ------------------------------------------------------- |
| `model`             | string           | The model to use for completion                         |
| `messages`          | array            | Array of message objects with `role` and `content`      |
| `temperature`       | number           | Sampling temperature between 0 and 2                    |
| `max_tokens`        | integer          | Maximum number of tokens to generate                    |
| `stream`            | boolean          | Stream the response using Server-Sent Events (SSE)      |
| `top_p`             | number           | Nucleus sampling probability mass                       |
| `frequency_penalty` | number           | Penalise repeated tokens based on frequency             |
| `presence_penalty`  | number           | Penalise tokens based on presence in prior text         |
| `tools`             | array            | List of tools the model may call                        |
| `tool_choice`       | string or object | Controls which tool (if any) the model uses             |
| `response_format`   | object           | Set to `{ "type": "json_object" }` to force JSON output |
| `stop`              | string or array  | Up to 4 sequences where the API will stop generating    |
| `seed`              | integer          | Seed for deterministic sampling                         |
| `logit_bias`        | object           | Modify the likelihood of specific tokens appearing      |
| `user`              | string           | A unique identifier for the end user                    |

## Streaming

Set `stream: true` to receive a streaming response using Server-Sent Events (SSE), identical to OpenAI's streaming format.

<CodeGroup>
  ```python Python theme={null}
  from openai import OpenAI

  client = OpenAI(
      api_key="YOUR_API_KEY",
      base_url="https://api.parallellabs.app/api/v0"
  )

  stream = client.chat.completions.create(
      model="gpt-4o",
      messages=[{"role": "user", "content": "Tell me about AI for business."}],
      stream=True,
  )

  for chunk in stream:
      if chunk.choices[0].delta.content is not None:
          print(chunk.choices[0].delta.content, end="", flush=True)
  ```

  ```javascript Node.js theme={null}
  import OpenAI from 'openai';

  const client = new OpenAI({
    apiKey: 'YOUR_API_KEY',
    baseURL: 'https://api.parallellabs.app/api/v0',
  });

  const stream = await client.chat.completions.create({
    model: 'gpt-4o',
    messages: [{ role: 'user', content: 'Tell me about AI for business.' }],
    stream: true,
  });

  for await (const chunk of stream) {
    process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
  }
  ```
</CodeGroup>

<Note>
  Each streamed chunk follows the standard OpenAI SSE format with `data:` prefixed JSON objects, ending with `data: [DONE]`.
</Note>

## Learn More

* [Chat Features Overview](/core-features/chat) — explore Parallel AI's chat capabilities
* [Chat Completions API Reference](/api-reference/introduction) — full endpoint documentation