Supported formats
Synjar supports common document formats for building your knowledge base.
Fully supported
PDF (.pdf)
- Text-based PDFs
- Complex layouts with tables and columns
- Multi-page documents
- Max size: 50 MB
note
Scanned PDFs (images) require OCR, which is not yet supported. Ensure your PDFs contain selectable text.
Microsoft Word (.docx)
- Word 2007 and later formats
- Formatted text, tables, lists
- Max size: 50 MB
Not supported: .doc (legacy format) - please convert to .docx
Plain text (.txt)
- UTF-8 encoded text files
- Simple, fast processing
- Max size: 10 MB
Markdown (.md)
- GitHub-flavored Markdown
- Headers, lists, code blocks
- Max size: 10 MB
Coming soon
We're working on support for:
- PowerPoint (.pptx)
- Excel (.xlsx)
- HTML files
- Images with OCR
Format recommendations
| Content type | Recommended format |
|---|---|
| Reports, manuals | |
| Procedures, policies | DOCX or MD |
| API documentation | Markdown |
| Simple notes | TXT |
Tips for best results
PDFs
- Use text-based PDFs, not scanned images
- Export from original software when possible
- Avoid password protection
Word documents
- Use Word's built-in heading styles
- Include a table of contents for long documents
- Save in modern .docx format
Markdown
- Use clear heading hierarchy (H1, H2, H3)
- Include descriptive link text
- Keep files focused on single topics