Gist

Last updated 2026-04-24 · 1 min read

PDF not working

PDFs work when they have a real text layer. Scanned-image PDFs, protected PDFs, and very large PDFs are the common failure modes. DOCX, PPTX, HTML, and other formats also upload directly.

PDFs come in three kinds. Which one yours is determines how well Gist reads it.

1. Text-layer PDF — works great

Created from a Word doc, a typeset paper, or a modern PDF export. You can select the text with your cursor and paste it elsewhere. Quality matches regular web articles.

2. Scanned-image PDF — hit or miss

A scan of a paper document is a series of images, even if it looks like text. If you cannot select any text in the PDF, it is a scanned-image PDF.

Gist can OCR these. Quality varies. A clean modern scan usually works. A low-resolution or skewed scan often does not.

3. Protected or gated PDF — fails

Passwords, DRM, or copy-locks block reading. Gist will say the file is not readable.

What to try

  • Select text in the PDF. You cannot? It is scanned — expect lower quality.
  • Try /try with a file upload. Drag the PDF in and let it run — the cleanest path for local PDFs.
  • Upload the original document instead. If you have the source file (Word, PowerPoint, HTML, etc.), upload that directly. Gist now accepts PDF, DOCX, XLSX, PPTX, HTML, CSV, TXT, Markdown, AsciiDoc, LaTeX, and XML.
  • Break up huge PDFs. Book-length PDFs can hit limits; try a chapter at a time via /try.
  • Export a cleaner version. If you control the PDF, re-export as a text-layer PDF and try again.

When to contact support

A text-layer PDF — one where you can clearly select text — failing or producing a wrong Gist is a bug. Send it via contact support:

  • The PDF itself (if you can share), or a link.
  • What you expected vs. what you saw.
  • Whether it is a paper, book chapter, report, etc.

What is next

Related