Fidelity is the point. The exam screen is deliberately plain — the real palette colours, the real layout, the real negative-marking display — because a mock that doesn't feel like the exam doesn't prepare you for it.
Answered, not answered, marked for review, not visited — the same colour-coded palette and states as the actual exam, not a simplified stand-in.
Jump between sections the way the real interface allows, with the palette and timer tracking each one separately.
The marks for a correct and an incorrect answer are shown on screen throughout — no surprise deduction at the end.
Flag a question and move on, then find every flagged question again from the palette before you submit.
Read a question in one language, answer it, switch language on the next — the interface tracks which language you actually read in.
Progress saves automatically as you work. Close the tab by accident and reopen the paper to pick up where you left off, with the clock counted honestly against the time you were away.
A test can mix all three in one paper — the interface adapts per question, not per test.
The standard four-option MCQ, scored the way every competitive exam does.
Select every correct option — full marks only if the selected set matches exactly, no partial credit.
Type the answer instead of picking an option, scored on an exact match.
Score and accuracy are the start, not the end. Every attempt gets a six-layer behavioural debrief — pacing, guessing discipline, navigation patterns, time-per-mark by section and more.
Five questions, the real interface, no account needed.