Files upload

Upload documents, images, and spreadsheets directly into a conversation. Le Chat reads the content and uses it as context, so you can ask questions, extract information, and get answers based on your data.

i
Information

File uploads are available on all plans with no setup required. Usage limits apply based on your plan. For details, visit our pricing page.

How to upload

How to upload

Two ways to get files into a conversation:

  • Drag and drop: drag one or more files into the chat window.
  • + button: click the + icon in the chat toolbar then select Upload Files and select files from your device.

You can upload multiple files at once (or even a folder). Le Chat keeps them in context for the full conversation, so you can ask follow-up questions without re-uploading.

tip

Use descriptive filenames. They make it easier to reference specific files when you have several in the same conversation.

Supported formats

Supported formats

CategoryFormats
DocumentsPDF, Word (.docx, .doc), PowerPoint (.pptx, .ppt), ODT, EPUB, RTF
SpreadsheetsExcel (.xlsx, .xls), CSV, ODS, Numbers
ImagesPNG, JPEG, WebP, GIF
Text and markupTXT, Markdown, RST, LaTeX
Data formatsJSON, JSONL, XML, YAML
Code filesPython, JavaScript/TypeScript, Java, Go, Rust, C/C++, Ruby, PHP, SQL, and more
EmailEML, MSG
Querying text documents

Querying text documents

Upload a PDF, Word file, or presentation and ask Le Chat to focus on what matters to you. Answers come directly from the uploaded content, not from general knowledge.

Typical prompts:

  • "Summarize this annual report with a focus on financial risks."
  • "Extract all dates, parties, and obligations from this contract."
  • "List the key security findings in this audit report."
  • "Compare slide 3 and slide 7 and list the differences."

Follow up to go deeper. Le Chat remembers uploaded files throughout the conversation, so you can refine, drill into specifics, or ask for translations without re-uploading.

Analyzing images

Analyzing images

Le Chat can interpret photos, diagrams, screenshots, and scanned pages. Upload an image with the + button and ask your question.

  • Text extraction: "Extract all the text from this scanned invoice." Works with handwritten notes, printed forms, and photographed whiteboards.
  • Diagram interpretation: "Describe the architecture shown in this diagram."
  • Screenshot analysis: "What error is shown in this screenshot?"
note

For heavier document processing needs (structured extraction from scanned PDFs, multi-page forms, or handwritten content in bulk), check Studio's Document Processing features.

Analyzing spreadsheets

Analyzing spreadsheets

For structured data like CSV or Excel files, enable Code Interpreter to get the most out of your uploads. Code Interpreter runs Python in a secure environment, so Le Chat can compute statistics, filter data, and generate charts directly in the conversation.

  1. Click the + icon or type / in the chat window.
  2. Select Tools then enable Code Interpreter.
  3. Upload your spreadsheet using the + button.
  4. Ask your question.

Typical prompts:

  • "Show me all sales from the North region."
  • "What's the average deal size by quarter?"
  • "Graph the monthly revenue trend as a line chart."

You don't need coding skills. Code Interpreter writes and runs the Python code for you, then displays results (tables, charts, statistics) inline.

Tips for better results

Tips for better results

  • Be specific. Tell Le Chat what to focus on: "Summarize this contract for a compliance officer" gives better results than "Summarize this file."
  • Use action verbs. Summarize, extract, compare, explain, highlight, list, convert.
  • Reference files explicitly. When you have several uploads, mention the filename: "In AnnualReport_2024.pdf, page 14, extract the revenue breakdown."
  • Iterate. Follow up with requests to refine, translate, or reformat the output.
  • Code Interpreter: run Python on your data for charts, statistics, and transformations.
  • Canvas: turn analysis results into polished documents, presentations, or code.
  • Open URL: analyze web pages by sharing URLs instead of uploading files.
  • Libraries: build permanent knowledge bases from your documents for use across conversations.