Name a topic — or hand it your own notes — and get a real test back. Sat, marked, and returned with feedback on the ones you missed.
1Which process converts light energy into chemical energy in plants?
2Chlorophyll absorbs mostly green light.
3Name the gas a plant takes in during photosynthesis.
It behaves like a test, not a stack of flashcards. Every question is on the page, you work through them in your own time, and nothing is marked until you submit — so there is no drip of little green ticks steering you as you go.
That matters more than it sounds. Immediate per-question feedback quietly turns recall practice into recognition practice, and recognition is the thing that feels like knowing and then deserts you in an exam hall.
Submit and you get a score, then a question-by-question breakdown: what you put, what the answer was, and — where you went wrong — feedback on the specific mistake rather than a restatement of the right answer.
Written answers are marked too, not skipped. Short and free-response questions get read and judged, and anything genuinely ambiguous is handed back flagged for your own review instead of being quietly guessed at.
Ask any generator for fifteen questions on one topic and it will cheerfully write you the same question fifteen ways — What is ½ + ¼? and then Find the sum of ½ and ¼, with different options and a different right answer, three questions apart.
Bonsai compares what each question actually asks — the claim, the numbers, the answer choices — and throws the repeats away before the test ever reaches you.
A fifteen-question test is fifteen questions. If a topic genuinely does not have fifteen distinct things in it, you get a shorter test and a note saying so, which is the honest answer.