# Reducto (/integrations/reducto)



{/* MANUAL-CONTENT-START:intro */}

Use [Reducto](https://reducto.ai/) to extract content from a PDF supplied as an upload or file reference. Optionally select 1-indexed pages and choose Markdown or HTML for table output. The result includes parsed chunks and blocks, job metadata, and usage information.

{/* MANUAL-CONTENT-END */}

## Use the block [#usage-instructions]

Extract text from uploaded PDF documents or file references.

## Actions [#actions]

### Reducto PDF Parser [#reducto-pdf-parser]

Parse PDF documents using Reducto OCR API

#### Input [#input]

| Parameter | Type | Required | Description |
| - | - | - | - |
| `file` | file | Yes | PDF document to be processed |
| `pages` | array | No | Specific pages to process (1-indexed page numbers) |
| `tableOutputFormat` | string | No | Table output format (`md` for Markdown or `html` for HTML). Defaults to `md`. |
| `apiKey` | string | Yes | Reducto API key (REDUCTO\_API\_KEY) |

#### Output [#output]

| Parameter | Type | Description |
| - | - | - |
| `job_id` | string | Unique identifier for the processing job |
| `duration` | number | Processing time in seconds |
| `usage` | json | Resource consumption data |
| `result` | json | Parsed document content with chunks and blocks |
| `pdf_url` | string | Storage URL of converted PDF |
| `studio_link` | string | Link to Reducto studio interface |