What is this feature?
ChatGPT can read images, charts, and diagrams embedded in PDFs uploaded to a chat. Users can upload a PDF, and ChatGPT can interpret the text and any visual elements within that file.
How does it work?
Select + in the prompt area, then Add photos & files to upload your PDF.
ChatGPT will read both the text and any embedded images or diagrams within the PDF.
You can then ask questions or request summaries—anything from extracting the main points of a report to explaining complex charts.
Is it compatible with GPTs and Projects?
Partially. PDFs uploaded as GPT Knowledge or Project Files are processed using text-only retrieval. PDFs uploaded by users during interactions with a published GPT or within a Project conversation are processed using visual retrieval.
Who can use it?
This capability is available on ChatGPT Plus, Pro, Business, Enterprise, and Edu accounts. It is not available to every Free account.
What problem does it solve?
You can ask questions about text, images, charts, and diagrams in a PDF without uploading each image separately.
Where can I find file-upload availability and limits?
For current availability and limits, see: Uploading files and audio to ChatGPT.