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

# Inferoute REST API Reference

> Complete reference for the Inferoute REST API. Route LLM requests across providers using an OpenAI-compatible interface at https://api.inferoute.ai/v1.

Inferoute exposes an OpenAI-compatible REST API that lets you send inference requests to multiple LLM providers through a single unified endpoint. Every request goes to `https://api.inferoute.ai/v1`, accepts JSON, and returns JSON — so any tooling that works with the OpenAI API works with Inferoute as well.

## Base URL

```bash theme={null}
https://api.inferoute.ai/v1
```

## Authentication

All requests require a Bearer token in the `Authorization` header:

```bash theme={null}
Authorization: Bearer YOUR_API_KEY
```

See [Authentication](/docs/api-reference/authentication) for full details and code examples.

## Available endpoints

| Method | Endpoint               | Description                                      |
| ------ | ---------------------- | ------------------------------------------------ |
| `POST` | `/v1/chat/completions` | Generate chat responses from a messages array    |
| `POST` | `/v1/completions`      | Generate text completions from a prompt string   |
| `POST` | `/v1/embeddings`       | Generate vector embeddings for text input        |
| `GET`  | `/v1/models`           | List all available models and their capabilities |

<CardGroup cols={2}>
  <Card title="Chat completions" icon="message" href="/docs/api-reference/chat-completions">
    Send a messages array and receive an assistant reply. Supports streaming, function calling, and all OpenAI-compatible parameters.
  </Card>

  <Card title="Completions" icon="text" href="/docs/api-reference/completions">
    Legacy text completion endpoint. Accepts a prompt string and returns generated text.
  </Card>

  <Card title="Embeddings" icon="vector-square" href="/docs/api-reference/embeddings">
    Generate vector representations of text for semantic search, RAG pipelines, and classification.
  </Card>

  <Card title="Models" icon="list" href="/docs/api-reference/models">
    Retrieve the full list of models available on Inferoute, including provider, context window, and capabilities.
  </Card>
</CardGroup>

## OpenAI SDK compatibility

Because Inferoute implements the OpenAI API spec, you can use any OpenAI SDK by pointing `base_url` at Inferoute:

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

  client = OpenAI(
      api_key="YOUR_INFEROUTE_API_KEY",
      base_url="https://api.inferoute.ai/v1",
  )

  response = client.chat.completions.create(
      model="openai/gpt-4o",
      messages=[{"role": "user", "content": "Hello!"}],
  )
  print(response.choices[0].message.content)
  ```

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

  const client = new OpenAI({
    apiKey: "YOUR_INFEROUTE_API_KEY",
    baseURL: "https://api.inferoute.ai/v1",
  });

  const response = await client.chat.completions.create({
    model: "openai/gpt-4o",
    messages: [{ role: "user", content: "Hello!" }],
  });
  console.log(response.choices[0].message.content);
  ```
</CodeGroup>

## Inferoute-specific headers

Inferoute extends the OpenAI request/response contract with a small set of headers for routing control and observability.

### Request headers

| Header                 | Values                                           | Description                                                   |
| ---------------------- | ------------------------------------------------ | ------------------------------------------------------------- |
| `X-Inferoute-Strategy` | `cost`, `latency`, `availability`, `round-robin` | Routing strategy for this request                             |
| `X-Inferoute-Fallback` | Comma-separated model IDs                        | Ordered list of fallback models if the primary is unavailable |

### Response headers

| Header                   | Description                                                        |
| ------------------------ | ------------------------------------------------------------------ |
| `X-Inferoute-Provider`   | The provider that served the request (e.g., `openai`, `anthropic`) |
| `X-Inferoute-Request-Id` | Unique request identifier for debugging and support                |
| `X-Inferoute-Latency-Ms` | End-to-end request latency in milliseconds                         |