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

# Boole AI REST API – Endpoints, Errors, and Base URLs

> Explore the Boole AI OpenAI-compatible REST API: base URLs for cloud and local, versioning, request and response formats, and all available endpoints.

The Boole AI API is a REST API that implements the OpenAI specification. All endpoints accept JSON and return JSON. You can use it with any HTTP client or OpenAI-compatible SDK — no custom library needed. Switch between the cloud API and a local binary deployment by changing a single base URL.

## Base URLs

| Environment  | Base URL                   |
| ------------ | -------------------------- |
| Cloud API    | `https://api.boole.dev/v1` |
| Local Binary | `http://localhost:8000/v1` |

## Authentication

All cloud API requests require an `Authorization` header with a Bearer token:

```
Authorization: Bearer YOUR_API_KEY
```

Local binary deployments do not require authentication. See the [Authentication](/api-reference/authentication) page for header examples, SDK setup, and key management.

## Request Format

* All text endpoints use `Content-Type: application/json`.
* Audio endpoints use `multipart/form-data`.

## Response Format

Successful responses return HTTP `200` with a JSON body. Error responses return a `4xx` or `5xx` status code alongside an `error` object:

```json theme={null}
{
  "error": {
    "message": "Invalid API key",
    "type": "authentication_error",
    "code": 401
  }
}
```

## Available Endpoints

| Endpoint                        | Description                                                        |
| ------------------------------- | ------------------------------------------------------------------ |
| `POST /v1/chat/completions`     | Chat completions with streaming, tool calls, and structured output |
| `POST /v1/completions`          | Legacy text completions                                            |
| `GET /v1/models`                | List available models                                              |
| `POST /v1/audio/transcriptions` | Transcribe audio using Whisper                                     |

## Error Codes

| HTTP Code | Type                    | Meaning                                                                                  |
| --------- | ----------------------- | ---------------------------------------------------------------------------------------- |
| `400`     | `invalid_request_error` | Malformed request or missing required field                                              |
| `401`     | `authentication_error`  | Missing or invalid API key                                                               |
| `403`     | `permission_error`      | API key does not have permission for this resource                                       |
| `429`     | `rate_limit_error`      | Rate limit exceeded (60 req/sec default)                                                 |
| `500`     | `internal_error`        | Server-side error — check [status.booleinference.com](https://status.booleinference.com) |

## Versioning

The API is versioned at `/v1`. Breaking changes will be introduced under a new version prefix (e.g. `/v2`) so that existing integrations continue to work without modification.

***

<CardGroup cols={2}>
  <Card title="Chat Completions" icon="message-lines" href="/api-reference/chat-completions">
    Generate chat responses with streaming, tool calls, and structured output.
  </Card>

  <Card title="Completions" icon="text" href="/api-reference/completions">
    Legacy text completion endpoint for prompt-in, completion-out workflows.
  </Card>

  <Card title="Models" icon="microchip" href="/api-reference/models">
    List all 42 available model slugs and their metadata.
  </Card>

  <Card title="Audio" icon="waveform-lines" href="/api-reference/audio">
    Transcribe audio files with Whisper Large v3.
  </Card>
</CardGroup>