X09 Document Intelligence · Included in Free
Chat with your PDFs and Word documents
Document Intelligence answers questions from the documents you upload and quotes the passages it used. It can summarize, pull out key points and action items, rebuild tables and extract fields into structured data.
What you can do
- Ask questions and get answers grounded in the document
- Summarize long documents and list key points and action items
- Extract tables as Markdown and fields as JSON
- Compare two documents and list the differences
- Work with PDF, DOCX, TXT, Markdown, CSV, JSON and HTML
How it works
- Upload a documentDrop in a PDF or Word file. It's converted to text on your device.
- Ask or pick a taskAsk a question, or choose summarize, extract, key points, tables or compare.
- Use the resultCopy it, export it or save it to a project.
Good to know
- Files are parsed on your device; only the extracted text is stored and sent when you run a task.
- Answers say when the document doesn't contain the answer, instead of guessing.
- Scanned PDFs without selectable text aren't supported yet. Use Image Studio's text extraction on a screenshot instead.
Questions
Can I chat with a PDF for free?
Yes. Document Intelligence is included on the Free plan, with limits on the number and size of documents.
Does it work with Word documents?
Yes. DOCX files work the same way as PDFs, along with text, Markdown, CSV, JSON and HTML.
Can it extract tables from a PDF?
Yes. It rebuilds tables as Markdown, and can extract specific fields into JSON.
Are my documents private?
Documents are parsed on your device and the text is stored in your account. It's sent to the AI only when you run a task, and isn't used to train models.