---
title: "What file formats and limits does PreciseMD support?"
description: "The extensions PreciseMD accepts, the difference between the OCR and conversion pipelines, and the size, page, and count limits in Playground and OCR tasks"
canonical_url: "https://docs.ardge.com/en/knowledge/precisemd/formats-and-limits"
---
# What file formats and limits does PreciseMD support?

> The extensions PreciseMD accepts, the difference between the OCR and conversion pipelines, and the size, page, and count limits in Playground and OCR tasks

## Short answer

PreciseMD runs OCR on eight extensions — PDF, JPG/JPEG, PNG, WEBP, BMP, and TIFF/TIF — and converts three Office formats, DOCX, XLSX, and PPTX. GIF is not supported. Files uploaded to **Playground** are capped at 100 MB per file and 100 pages per PDF; an **OCR task** reading source files from a storage connection has no page limit and a 150 MB per-file cap.

## Details

### Supported formats

The upload dialogs throughout the interface state "Supports PDF, JPG, PNG, WEBP, BMP, TIFF, DOCX, XLSX, PPTX". The extensions actually accepted are:

<table>
<thead>
  <tr>
    <th>
      Format
    </th>
    
    <th>
      How it's processed
    </th>
    
    <th>
      Notes
    </th>
  </tr>
</thead>

<tbody>
  <tr>
    <td>
      PDF
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Split by page and parsed page by page
    </td>
  </tr>
  
  <tr>
    <td>
      JPG, JPEG
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Treated as a single-page document
    </td>
  </tr>
  
  <tr>
    <td>
      PNG
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Treated as a single-page document
    </td>
  </tr>
  
  <tr>
    <td>
      WEBP
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Treated as a single-page document
    </td>
  </tr>
  
  <tr>
    <td>
      BMP
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Treated as a single-page document
    </td>
  </tr>
  
  <tr>
    <td>
      TIFF, TIF
    </td>
    
    <td>
      OCR pipeline
    </td>
    
    <td>
      Treated as a single-page document
    </td>
  </tr>
  
  <tr>
    <td>
      DOCX
    </td>
    
    <td>
      Conversion pipeline
    </td>
    
    <td>
      Converted straight to Markdown, no OCR
    </td>
  </tr>
  
  <tr>
    <td>
      XLSX
    </td>
    
    <td>
      Conversion pipeline
    </td>
    
    <td>
      Converted straight to Markdown, no OCR
    </td>
  </tr>
  
  <tr>
    <td>
      PPTX
    </td>
    
    <td>
      Conversion pipeline
    </td>
    
    <td>
      Converted straight to Markdown, no OCR
    </td>
  </tr>
</tbody>
</table>

<warning>

**GIF is not supported:**

GIF is excluded from the OCR pipeline because the tomography service can't decode it. Uploading one shows "This file format isn't supported. Upload a PDF, an image, or an Office document instead".

</warning>

### Count and size limits

<table>
<thead>
  <tr>
    <th>
      Limit
    </th>
    
    <th>
      Value
    </th>
    
    <th>
      Message when exceeded
    </th>
  </tr>
</thead>

<tbody>
  <tr>
    <td>
      Per-file size uploaded to <strong>
        Playground
      </strong>
    </td>
    
    <td>
      100 MB
    </td>
    
    <td>
      <code>
        The file exceeds the 100 MB limit. Compress it and upload again
      </code>
    </td>
  </tr>
  
  <tr>
    <td>
      PDF pages uploaded to <strong>
        Playground
      </strong>
    </td>
    
    <td>
      100 pages
    </td>
    
    <td>
      <code>
        The PDF exceeds the 100-page limit. Trim it and upload again
      </code>
    </td>
  </tr>
  
  <tr>
    <td>
      Per-file size read by an <strong>
        OCR task
      </strong>
    </td>
    
    <td>
      150 MB
    </td>
    
    <td>
      —
    </td>
  </tr>
  
  <tr>
    <td>
      Files per upload in <strong>
        Playground
      </strong>
    </td>
    
    <td>
      50
    </td>
    
    <td>
      <code>
        You can upload at most 50 files
      </code>
    </td>
  </tr>
  
  <tr>
    <td>
      Files per upload to a <strong>
        storage connection
      </strong>
    </td>
    
    <td>
      500
    </td>
    
    <td>
      —
    </td>
  </tr>
</tbody>
</table>

The 100 MB and 100-page limits are validated at upload time, so they apply only to files uploaded through the interface. An **OCR task** reads existing files straight from a storage connection without that validation, which is why it has no page limit.

### What happens to unsupported formats

When an **OCR task** scans the source folder, it skips files whose extension isn't on the supported list — they aren't reported as failures. To process a GIF, save it as a PNG or JPG at the source before putting it in the source folder.

## How to check

To confirm whether a given file will be accepted:

1. Go to the **Playground** page
2. Click **Upload** in the *Source documents* panel
3. Drag the file onto the dialog, or click it to select the file

The bottom of the dialog shows "N / 50 files selected". Unsupported formats can't be added to the selection; files over the size or page cap show the matching error after you submit.

## Related articles

- [What files are produced in the OCR output folder?](https://docs.ardge.com/en/knowledge/precisemd/ocr-output-files)
- [Tune prompts and compare OCR results in Playground](https://docs.ardge.com/en/knowledge/precisemd/playground-prompt-tuning)


## Sitemap

See the full [sitemap](https://docs.ardge.com/sitemap.md) for all pages.