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

# Supported Formats

> Document types you can upload to Finch Fusion

# Supported Formats

Finch Fusion supports a variety of document formats for upload and analysis. This guide covers supported file types and their characteristics.

## Fully Supported Formats

### PDF Documents (.pdf)

PDF is the most common format for business and financial documents.

**Supported features:**

* Native text extraction
* OCR for scanned documents
* Multi-page documents
* Embedded images and tables

**Best for:**

* Annual reports (10-K, 10-Q filings)
* Research papers
* Presentations converted to PDF
* Scanned documents

<Tip>
  For best results, use text-based PDFs rather than scanned images when possible. Text-based PDFs process faster and more accurately.
</Tip>

### Word Documents (.docx)

Microsoft Word documents are fully supported.

**Supported features:**

* Full text extraction
* Formatting preservation for analysis
* Tables and lists
* Embedded images (text from images may not be extracted)

**Best for:**

* Draft reports
* Internal memos
* Written analyses
* Meeting notes

<Info>
  Older `.doc` format may have limited support. Convert to `.docx` for best results.
</Info>

### Plain Text Files (.txt)

Simple text files are processed quickly and accurately.

**Supported features:**

* Full text extraction
* Fast processing
* Universal compatibility

**Best for:**

* Data exports
* Transcripts
* Simple notes
* Log files

## Format Comparison

| Format           | Text Accuracy | Processing Speed | Best Use Case                  |
| ---------------- | ------------- | ---------------- | ------------------------------ |
| PDF (text-based) | Excellent     | Fast             | Official reports, filings      |
| PDF (scanned)    | Good          | Slower           | Physical document digitization |
| DOCX             | Excellent     | Fast             | Draft documents, notes         |
| TXT              | Perfect       | Very Fast        | Plain text content             |

## File Size Considerations

### Recommended File Sizes

| File Type | Recommended Max | Notes                                   |
| --------- | --------------- | --------------------------------------- |
| PDF       | 50 MB           | Larger files take longer to process     |
| DOCX      | 25 MB           | Images increase file size significantly |
| TXT       | 10 MB           | Very large text files may timeout       |

### Handling Large Documents

If your document exceeds recommended sizes:

1. **Split the document** - Break it into logical sections
2. **Compress images** - Reduce embedded image sizes
3. **Remove unnecessary pages** - Include only relevant content

## Optimizing Documents for Upload

### PDF Best Practices

* Use text-based PDFs when possible
* Ensure scanned documents are at least 300 DPI
* Flatten complex PDFs before upload
* Remove password protection before uploading

### Word Document Best Practices

* Save as `.docx` format (not `.doc`)
* Compress images before adding to the document
* Remove track changes and comments
* Ensure fonts are embedded or use standard fonts

### General Tips

* Give files descriptive names
* Ensure documents are not corrupted before upload
* Remove any password protection
* Verify the document opens correctly on your computer first

## Unsupported Formats

The following formats are **not currently supported**:

* Spreadsheets (`.xlsx`, `.xls`, `.csv`)
* Presentations (`.pptx`, `.ppt`)
* Images (`.jpg`, `.png`, `.gif`)
* Web pages (`.html`)
* Email files (`.eml`, `.msg`)
* Compressed archives (`.zip`, `.rar`)

<Info>
  For unsupported formats, consider converting to PDF or copy-pasting content into a text file before upload.
</Info>

## Format Conversion Tips

If you need to convert documents:

### Spreadsheets to PDF

1. Open in Excel or Google Sheets
2. Use **Print** > **Save as PDF**
3. Upload the resulting PDF

### Presentations to PDF

1. Open in PowerPoint or Google Slides
2. Use **File** > **Export as PDF**
3. Upload the PDF

### Web Pages to PDF

1. Open the page in your browser
2. Use **Print** > **Save as PDF**
3. Upload the PDF

## Troubleshooting Format Issues

<AccordionGroup>
  <Accordion title="My PDF isn't processing correctly">
    The PDF may be image-based without embedded text. Finch Fusion uses OCR for scanned documents, but results depend on image quality. Consider using a higher-resolution scan.
  </Accordion>

  <Accordion title="Word document formatting looks wrong">
    Formatting is used for analysis but not preserved in display. Focus on the text content being extracted correctly rather than visual formatting.
  </Accordion>

  <Accordion title="My file type isn't supported">
    Convert the file to a supported format (preferably PDF) before uploading. Most applications have Export as PDF functionality.
  </Accordion>
</AccordionGroup>