A PDF to quiz tool
that turns a chapter into four-option questions.
Drop the PDF, the extracted text stays in your browser, and every question shows the line it was built from.
No account, two generations a day, files up to 15MB and 25,000 characters free.
Open any question to see the line it came from. Paste 40 characters or more to start.
PDF to quiz vs PDF to flashcards: which fits your revision
Both modes read the identical extraction from the identical upload, so the choice is not about the PDF, it is about which kind of check you want. Quiz mode hands you four options and asks you to recognize the right one, which is closer to what an actual multiple choice exam will ask of you and is the faster of the two to work through a whole chapter with.
Flashcards ask for recall instead: no options on screen, just the question, and you say the answer before you flip it. That is the harder test, and it is usually the more honest one, because picking the right answer out of four is not the same skill as producing it from nothing. A dense chapter you have already read once is a reasonable candidate for flashcards; a chapter you are seeing for the first time tonight is often better as a quiz, since recognition at least confirms you followed the material at all.
You do not have to commit to one. Switch the tab above at any point and the same extracted text regenerates in the other mode, so trying quiz first and flashcards second on the same upload costs nothing but a second click.
What makes a PDF chapter quiz well, and what doesn't
With typed notes you can rewrite a thin bullet into a proper sentence before pasting it in. A PDF does not give you that option: whatever the document's own author wrote is what gets extracted, bullets and all, and the quiz can only build a question from a claim that is actually there.
A textbook chapter or a journal PDF written in prose tends to quiz well for exactly this reason: full sentences already state a subject, a condition and an outcome, which is the shape a multiple choice question is built from. A PDF of lecture slides, on the other hand, is often just short phrases and headers with the actual explanation delivered out loud, live, and never written down. Extracting that gives the generator less to work with, and the questions that come back will read thinner, not because of a bug, but because the source material was never a full sentence to begin with.
If a slide-deck PDF is what you have, the fix is not this tool. It is finding the lecture notes, the textbook chapter, or your own written-up summary that covers the same material in sentences, and uploading that instead.
The PDF stays in your browser; the generator only sees the text it pulls out
Extraction happens client-side, in the tab you are already looking at. The file is read into memory, its text layer is pulled out page by page, and that text fills the paste box. Only once you press generate does anything leave your browser, and what leaves is the extracted text sitting in that box, not the PDF file itself.
Open your browser's devtools, switch to the network panel, and drop a PDF onto the page: no upload request fires until you click generate, and even then there is no PDF in the request, only the text. For a lecture PDF with a professor's name on it or unpublished course material, that distinction is the whole reason to check rather than take it on faith.
Where this stops
No OCR, so a scanned or photographed PDF returns the "no text found" message instead of a guess at what the page might say. The free tier reads the first 25,000 characters of whatever the PDF extracts to; a longer chapter is trimmed with a visible note, and reading a whole long PDF end to end is a Pro waitlist feature, not something available today. Multiple choice only in quiz mode, four options and a source line, no short answer grading. What is left, reading the PDF accurately and showing where each question came from, is the part we tried to get right.
Before you turn a PDF into a quiz
How many quiz questions come out of one PDF chapter?
As many as the chapter gives the model distinct claims to work with, not a count tied to page number or file size. A dense textbook chapter written in full sentences can return a dozen questions; the same page count from a slide-deck PDF, which is mostly short bullet fragments once extracted, often returns half that, because a fragment does not carry enough of a claim to build a fair multiple choice question from.
Does it write multiple choice straight from the PDF, or only from text I type?
From the PDF. Drop the file in and the browser extracts its text layer first; that extracted text is what goes to the generator, the same as if you had typed or pasted it yourself. Nothing about the question-writing step treats a PDF's text any differently once it is out of the file.
My PDF is a scan. Why did nothing come back?
A scanned page is an image of text, not text, and this tool reads a PDF's text layer rather than running OCR on it. Try Ctrl+F on the PDF in your regular reader: if it cannot find a word you can see on the page, there is no text layer for this tool to read either, and it says so instead of returning a guess.
Can I check a question against the actual PDF line it came from?
Open the question and it highlights the sentence inside the extracted text, the same text that filled the paste box after upload. That is the PDF's own wording, not a paraphrase, though what you are checking against is the extracted text, not a rendering of the PDF's page image.
What's the free limit for a PDF turned into a quiz?
15MB per file, and 25,000 characters of extracted text, roughly a chapter. A longer PDF is trimmed to that many characters with a note saying so, rather than silently cutting off partway through a sentence. Reading a whole long PDF is on the Pro waitlist, not something the free tier does today.
Can I export the quiz once it's built from a PDF?
Yes, the same Export for Anki and Export CSV buttons that show up after any generation, quiz or flashcards. A question becomes the front, its correct answer becomes the back; the other three options are not part of either export, since Anki and a spreadsheet both expect a front and a back, not a four-way choice.
Other ways into the same generator
Or start from the QuizPaste home page, where the same tool takes any text you paste.
QuizPaste is not affiliated with Adobe, Anki or AnkiWeb. PDF is a format, not a trademark we are claiming any relationship to; Anki is named only to describe the file format our export is compatible with.