> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/lumina-ai-inc/chunkr/llms.txt
> Use this file to discover all available pages before exploring further.

# API Overview

> Introduction to the Chunkr API for document layout analysis and chunking

The Chunkr API is a powerful document processing service that converts documents into RAG/LLM-ready data through advanced layout analysis and intelligent chunking.

## Base URL

All API requests should be made to:

```
https://api.chunkr.ai
```

## API Versioning

The current API version is **v1**. All endpoints are prefixed with `/api/v1`.

Example endpoint:

```
https://api.chunkr.ai/api/v1/task
```

## Request Format

The API accepts requests in two formats:

* **JSON** (Recommended): `Content-Type: application/json`
* **Multipart Form Data**: `Content-Type: multipart/form-data` (deprecated)

## Response Format

All successful responses return JSON with appropriate HTTP status codes:

* `200 OK` - Request successful
* `400 Bad Request` - Invalid request parameters
* `401 Unauthorized` - Authentication failed
* `404 Not Found` - Resource not found
* `429 Too Many Requests` - Rate limit exceeded
* `500 Internal Server Error` - Server error

## Rate Limits

<Note>
  The API implements intelligent rate limiting to ensure fair usage and optimal performance.
</Note>

Chunkr uses a token bucket algorithm for rate limiting across different service types:

### Service Rate Limits

| Service            | Default Rate Limit | Configurable |
| ------------------ | ------------------ | ------------ |
| **General OCR**    | 5 requests/second  | Yes          |
| **Segmentation**   | 5 requests/second  | Yes          |
| **LLM Processing** | Varies by model    | Yes          |

### Batch Sizes

To optimize throughput, the API processes requests in batches:

* **General OCR**: 30 pages per batch
* **Segmentation**: 3 pages per batch

### Rate Limit Headers

Currently, rate limit information is managed server-side. If you exceed the rate limit, you'll receive a `429 Too Many Requests` response.

<Warning>
  When you hit a rate limit, the API will return a 429 status code with the message "Usage limit exceeded". Implement exponential backoff in your retry logic.
</Warning>

## Timeouts

Different operations have different timeout configurations:

* **General OCR**: Configurable (no default timeout)
* **Segmentation**: Configurable (no default timeout)
* **LLM Processing**: 150 seconds default
* **API Request**: 600 seconds (10 minutes)

## File Size Limits

<Note>
  The API accepts files up to **1 GB** by default. Both total request size and in-memory limits are enforced.
</Note>

* **Max Total Limit**: 1 GB (configurable via `MAX_TOTAL_LIMIT`)
* **Max Memory Limit**: 1 GB (configurable via `MAX_MEMORY_LIMIT`)

## Supported File Types

The API automatically detects file types and supports various document formats. Common MIME types include:

* PDF documents
* Images (JPEG, PNG)
* Other document formats

## Health Check

Check the API health and version:

```bash theme={null}
curl https://api.chunkr.ai/health
```

Response:

```
OK - Version {version}
```

## API Documentation

Interactive API documentation is available at:

* **Swagger UI**: `https://api.chunkr.ai/swagger-ui/`
* **ReDoc**: `https://api.chunkr.ai/redoc`
* **OpenAPI Spec**: `https://api.chunkr.ai/openapi.json`

## Next Steps

<CardGroup cols={2}>
  <Card title="Authentication" icon="key" href="/api/authentication">
    Learn how to authenticate your API requests
  </Card>

  <Card title="Error Handling" icon="triangle-exclamation" href="/api/errors">
    Understand error responses and status codes
  </Card>

  <Card title="Task Management" icon="list-check" href="/api/tasks/create">
    Create and manage document processing tasks
  </Card>
</CardGroup>