Anything can produce four options and a tick. The difficulty is writing three wrong ones that a half-prepared student would genuinely fall for, and that is what this is built around.
Bigger sets read more of your document. Included with Student and Genius.
PDF, PowerPoint, Word, or a link · up to 10 MB · start without an account
Upload a file, paste a link, or paste the text. The questions can only be about what you supply.
Each question gets one defensible answer and three distractors drawn from the same material, then the set is screened for the usual giveaways.
Every option comes back explained, so a wrong pick tells you which misconception you were holding.
Three lines of statistics revision, and the marked paper showing how the wrong options were built. Sample material, written by us to show the shape of the output.
What you supplied
Hypothesis testing notes, pasted
Your marked paper
Your score
7 of 10
3 of the 10 questions
A study rejects a null hypothesis that was in fact true. Which error has occurred?
Increasing the sample size raises which quantity directly?
Why is power written as one minus beta rather than as its own parameter?
Power is the probability of correctly rejecting a false null, which is exactly the complement of failing to do so. The distractors here were the other three quantities in your notes, which is why guessing did not help.
The three you missed come back on their own as a shorter round, and redrilling them spends no part of your allowance.
The scored multiple choice quiz needs no payment, so this page describes the Free plan in full. Student and Genius add flashcards and written recall over the identical question set.
Sets are checked before you see them for the patterns that let people score without knowing anything: an answer that gives itself away by being the longest, two options that say the same thing, and "all of the above".
The wrong options are drawn from neighbouring ideas in the document rather than invented, so they are the confusions your subject actually produces.
A question with two arguably correct options teaches you to distrust the test. Each stem is written so exactly one option survives the material it came from.
Multiple choice is recognition. When you want production instead, a paid plan runs the same questions as written recall with nothing to choose between.
Sitting a multiple choice paper soon? Read our guide to multiple choice test strategy.
A good wrong option is one a student who half understands the material would choose for a reason. That usually means it is a neighbouring concept, a common misapplication, or the right answer to a slightly different question, and it means it is roughly the same length and register as the correct option. Bad distractors fail in predictable ways: they are obviously absurd, they are much shorter than the answer, they contradict each other so that two can be eliminated at once, or they are so vague that nothing rules them out. The reason this matters beyond neatness is that a test you can pass by elimination measures your test technique rather than your knowledge, and revising against it builds confidence that the real paper then removes. The screening pass exists precisely because a generator left to itself drifts toward those patterns, since they are the easiest options to produce.
Each set is 10 questions on the Free plan, rising to 30 on a paid one. The larger sizes are spread over proportionally more of your document rather than asking about the same passages repeatedly, so a bigger set means wider coverage rather than denser interrogation of the opening pages. 10 is chosen deliberately as a size you will finish in one sitting, because a long paper generated late at night reliably gets abandoned somewhere around question twenty and teaches nothing. To go further through the material, upload the next section. Uploading the same document again does not restart it, since the material is recognised by its text and generation continues into the part you have not been tested on yet, which is what lets a long document be worked through across several sittings rather than sampled repeatedly from the front.
Questions are not editable, and that is a real limitation rather than an oversight, so it is worth weighing before you rely on this. What you can do is flag a question you think is wrong, which is the mechanism that exists for a bad item. If you need a bank you can revise, reword, and hand out, a purpose-built assessment tool will serve you better than this will. What this is built for is a single student testing themselves quickly on their own material, where the value is in getting 10 defensible questions in under a minute rather than in curating them afterwards. The practical answer for a question that reads oddly is to check it against your source, since it can only have come from something in the document you supplied.
It is the one mode with no paywall anywhere near it. Scored multiple choice runs on every tier, so a Free account produces 2 complete question sets from your own material, screened, marked, explained, and redrillable, without a card. The cap is on how many sets you generate rather than on what a set contains, and those 2 arrive once for the life of the account instead of resetting each month. Beyond them, Student is $14.99 monthly for 100 generations, and Genius is $59.99 for a year against a fair-use ceiling of 1000 a month. Paying widens a set to 30 and opens two further ways to study the same questions, but nothing about the multiple choice generator itself improves when you pay, because there is no reduced version of it to improve on.
Most tools on this SERP will produce options. We looked at how good they are: the best AI tools for turning notes into a quiz, and the best AI tools for exam prep.