Working with files
PDF, DOCX, XLSX, and images
You can attach PDF, DOCX, XLSX, and images to your messages. The AI reads the content and answers questions about it. Specify which file or section you're asking about if you have several attachments.
For PDF and DOCX, ask for summaries, key points, or specific information. For XLSX, you can ask to edit cells, add rows, create tables, or perform calculations. The AI will use special tools to work with spreadsheets.
When asking about a file, give context: "In the attached report, what are the main conclusions?" or "Add a row to the table with the following data." The clearer the request, the more accurate the result.
Large files may take longer to process. For very long documents, ask about specific sections or chapters. "Summarize chapter 3" works better than "Summarize the entire 200-page document" in one go.
For XLSX, the AI can read, edit cells, add rows, and finalize changes. Describe the edits clearly: "Change cell B5 to 42" or "Add a row at the end with Name: John, Age: 30." After edits, the model will offer to save the file.