Loading…
30 original dMAT-style questions across all four subtests, timed at 30 minutes, with a worked explanation on every one. Not official dMAT questions — no official paper has ever been published.
Free — it just needs an account, which takes a moment.
The 30 questions are spread across the four task types the dMAT actually sets, so nothing in it is a surprise on the day. Each one is written to the official format rather than adapted from a general aptitude book, which matters more than it sounds: the reasoning a CAT or GRE question asks for is not the reasoning these ask for.
Three of those four sit inside the Core Module, which measures general study aptitude through Figure Sequences, Mathematical Equations and Latin Squares. The fourth, General Academic, is the Subject Module that applicants going through APS India take. If the structure is still new to you, the dMAT syllabus and exam pattern sets out both modules in order, and what the dMAT is explains where the test sits in the application process before you spend an evening preparing for it.
One clock runs across the whole mock: 30 questions in 30 minutes, which works out at about a minute each. The deadline is set on the server when the attempt begins, so refreshing the page, closing the tab or coming back on another device does not buy you time — the clock you see is the clock that is actually running.
Inside the mock you move with Previous and Next, jump to any question from the navigator strip, and flag anything you want to revisit. A red banner warns you once under two minutes remain. When the clock reaches zero the attempt submits itself and you go straight to your results, with anything you never reached marked unanswered. If you have to stop halfway, returning to the mock resumes the attempt you already started rather than throwing your answers away.
That per-question budget is close to the real thing on purpose. Each Core subtest sets roughly twenty tasks in about twenty-five minutes, and the General Academic Subject Module runs about ninety minutes, with the whole examination taking around three and a half hours including the break between modules. A minute a question is the tempo the Core asks for, and half an hour is long enough to find out whether you can hold it.
Pacing is a skill, not a temperament. Work out your per-question budget before you start, decide in advance what you will do when a question clearly overruns it — flag it, take your best current option, move on — and protect the easy marks further down rather than sinking three minutes into one hard item. More on that in dMAT time management and pacing.
On a first attempt most lost marks are not lost to weak reasoning. They go to the format — a clock that runs down while you are still reading the instructions, a grid you would have solved on paper but have to hold in your head, an answer you would have spotted by working backwards from the options if the options were there. Rehearsing the format is the part that transfers, and it is the only part you can rehearse: because no past papers exist, there is nothing to memorise even if you wanted to.
Thirty minutes is short on purpose. It is long enough to find out whether the pacing is a problem for you and short enough that you will actually finish it in one sitting.
Most preparation happens on paper. You print a set, spread it across the desk, circle the promising options, scribble a step in the margin and strike through what you have ruled out. None of that is available in the exam room: the questions are on a monitor, the answers are selected with a mouse, and the Core subtests are designed to be solved mentally. A practice score earned with a pad beside you is measuring a different candidate.
So take this mock the way the exam will be taken. Use a real computer rather than a phone, which reflows text and shows you one element at a time. Read at full size and work the question in front of you instead of scrolling back and forth. Do not annotate anything. Keep the desk clear of notes and formula sheets, because they are prohibited in the exam room. And sit it in one go, without pausing to look something up — a comfortable practice run quietly overstates how ready you are.
Working on a screen is mostly a problem of keeping your place, and the fix is structure rather than effort. Scan in a fixed order every time so you never have to re-check whether you missed something. Use the cursor as a pointer to hold a row, a cell or an option. Name rules out loud in your head — “the inner shape rotates, the outer one stays” — because a rule you can say survives far better than a shape you are trying to picture, and keep the set of options still standing small enough to carry. Our guides to practising for the dMAT on a screen and to simulating real exam conditions cover the medium and the routine around it respectively.
The moment you finish you get a score out of 30, the percentage that represents, and a bar for each of the four task types showing how many you got right out of how many you attempted. The subtest breakdown is the useful part: it turns “I should practise more” into a specific thing to practise, and it is usually the first thing that tells you your problem is one task type rather than the paper as a whole.
Below that, every question comes back with the answer you chose, the correct answer, and a worked explanation for each one you attempted — the reasoning written out rather than a bare letter, including why the tempting wrong option was put there. Questions you never reached are shown separately as unanswered, which keeps a pacing failure from being filed as a knowledge gap.
One thing the report deliberately does not do is pretend to be official. What you get is raw accuracy, timing and task-level performance. The official dMAT certificate reports each module on a scale from 0 to 200 together with a percentile rank that places you against other participants, and no unofficial platform can reproduce that standardisation without official data — which is why you should be wary of any provider quoting you a predicted dMAT score. The dMAT score, percentile rank and certificate explained sets out what the official numbers mean and how they differ from a practice percentage.
A mock is not valuable because you completed it. It is valuable because of what you change afterwards, and the most common way to waste one is to glance at the score, skim two explanations and start another. Give the review at least as much attention as the attempt.
Start by sorting the questions into four groups rather than two: correct and efficient, correct but slow, incorrect, and guessed or low-confidence. The second and fourth groups are the ones that predict your next bad paper, and neither of them shows up in the total. A question you got right in three minutes is a question you will get wrong when the clock is tighter.
Then name the cause of each loss, and be specific about it. Avoid the word “careless”: it is a feeling, not a diagnosis, and it tells you nothing you can practise. Every cause has a drill attached to it.
| What went wrong | What to practise next |
|---|---|
| Method gap — you did not know how to start | Relearn the first move for that task type, then do five questions of it untimed |
| Interpretation error — you misread a phrase, unit or constraint | Translate ten prompts into equations or rules without solving any of them |
| Execution error — right method, wrong arithmetic or placement | Slow the working to one line per step and back-substitute every answer |
| Pacing error — you overran, or rushed and lost easy marks | Run short timed sets of five to ten questions with a fixed abandon point |
| Interface error — you mis-selected, mis-flagged or lost your place | Take the next set on screen only, using the navigator and flag deliberately |
| Confidence error — you changed a right answer, or guessed too surely | Mark your confidence before submitting each answer, then score the marks |
Reattempt in two stages. Close the explanation and solve the question again from the beginning while it is fresh, then come back to it several days later and explain the method before you choose an answer. The delayed attempt is the one that tells you whether you learnt a process or merely remembered an option. Keep the recurring causes in one place so patterns become visible across papers — the same error appearing three times is worth more of your week than any single hard question. How to review a dMAT mock test properly gives the full routine, and keeping a dMAT error log covers what to write down.
The four task types look unrelated and reward different first moves. Once the mock has told you which one is costing you, the work is specific.
You are not trying to see the answer. You are reducing a busy picture to a small number of testable changes. Run a fixed property checklist — count, position, orientation, reflection, shading, size, combination, and whether two rules alternate — and start with the easiest visible property rather than inventing a complicated rule. Compare adjacent cells first, describe the change in one sentence, then check whether that same sentence explains the next transition. Test rows and columns separately instead of assuming the direction, track independent layers one at a time, and state what the missing figure should contain before you look at the options. If a rule needs an “except in this cell”, it is the wrong rule. A step-by-step framework for dMAT figure sequences works the whole method through, and you can drill the format on the figure sequences practice page.
This subtest does not test advanced mathematics. It tests whether you can reason clearly about simple relationships while a clock runs. Read the whole system before touching any numbers, because the fastest solve almost always comes from spotting a relationship between the equations rather than grinding through one of them. Then choose your move deliberately: add when a variable carries opposite signs, subtract when a coefficient matches, substitute when one equation already gives a variable cleanly. Deciding before you write prevents the classic time sink of starting to substitute and then restarting. Work in small checkable steps, and drop your value back into the simplest original equation as a one-line check — that check catches most sign slips. How to solve dMAT mathematical equations under time goes further, and the mathematical equations practice page has the format untimed.
A grid where every letter appears once per row and once per column is a controlled elimination problem, not a puzzle to stare at. Begin with the row or column that has the fewest blanks. For any empty cell, the valid candidates are the intersection of what is still missing from its row and what is still missing from its column — that one principle resolves a large share of deductions on its own. Keep possibilities and certainties separate, because placing a letter merely because it is possible corrupts everything after it. When no row or column is nearly full, flip the question: instead of asking what belongs in this blank, ask where a particular letter can still go. And remember the answer options are constraints too — an option that duplicates a letter in the marked row or column is gone without any further work, and if the question asks for one cell you should stop the moment that cell is proven. How to solve dMAT-style Latin squares systematically sets out the full scan order, and the Latin squares practice page lets you build the habit slowly.
The Subject Module asks you to work with a short academic text, table or figure and answer questions about what is actually stated. The surface topic will often be unfamiliar and that is the point: the skill is using the information provided rather than recalling anything. Use the same four stages every time — identify what the question wants, locate the specific evidence, apply the reasoning, then verify that your answer uses the right unit, answers the exact question asked and does not go further than the evidence allows. Read purposefully: question stem first, then scan for the relevant section, then read that closely. With data, check the title, axes, units and whether the figures are totals, averages or percentages before you calculate anything. Be careful with absolute words — always, proves, causes, only — because an option can use the passage's own vocabulary while changing the relationship between the ideas. How to prepare for the dMAT General Academic Module covers the routine in detail, and the General Academic practice page has passage-based sets.
Sit one early, before you feel ready. The instinct is to cover the material first and keep the mock for when you have earned it, which means the single most informative thing you can do arrives after most of your decisions have already been made. An early paper tells you how long a question actually takes you rather than how long you assume it takes, and whether your concentration survives a full sitting — neither of which is visible while you work through topics at your own pace.
After that, count review cycles rather than mocks. A paper followed by a thorough review is one learning cycle; two papers in two days is two tests and no cycles at all. For most people one full mock every seven to ten days works through the main body of preparation, with shorter timed sets in between, and two or three in the final fortnight when the goal shifts from learning material to rehearsing conditions. The honest test of your frequency is whether you can name one specific thing you will do differently after each attempt.
Expect the numbers to move around. On a test this short a couple of lucky guesses and one misread question is a three-question swing with nothing behind it, so judge yourself on a band drawn from several attempts rather than on your most recent one, and hold the conditions steady between papers so the comparison means something. Our dMAT preparation guide lays out a week-by-week plan around all of this, and how often to take dMAT mock tests works through the spacing.
From here, the free practice questions let you work through the formats one at a time without a clock, and the free diagnostic does the same job as this mock but saves the result to your account so you can measure yourself against it later.
It is not an official g.a.s.t. paper and it does not contain official dMAT content. No official dMAT paper has been released, which means every “dMAT past paper” on the internet is either invented or lifted from a different exam — and a question written for a different exam trains the wrong reflex. What these questions do is match the published task formats: the same structures, the same constraints, the same single-choice answer types, so that the format itself is boring by the time it counts.
It is also not a predictor. Use the result diagnostically — which task types produced errors, whether time pressure was the real issue, which correct answers were guesses, whether repeated mistakes are decreasing — and leave the forecasting alone.
The first dMAT in India takes place on 26 September 2026. Registration opened on 29 June 2026 and closed on 15 September 2026, and certificates become available on 12 October 2026. The fee is €150. It is a digital, single-choice examination taken in English at licensed g.a.s.t. test centres, of which Ahmedabad is one, and it runs as a Core Module of three subtests — roughly twenty tasks in about twenty-five minutes each — followed by a General Academic Subject Module of about ninety minutes, for a total of around three and a half hours including the break between modules. Each module is reported on a scale from 0 to 200, along with a percentile rank.
The requirement applies to Master's applicants from the Summer Semester 2027 intake onward, and a transitional exemption relates to APS registrations completed before 29 June 2026. Whether it applies to you depends on your own timeline, so read what the dMAT is and who it affects and confirm everything against the official sources below — dates, fees and requirements are set by g.a.s.t. and APS India and can change.
Exam facts on this page last verified against official sources on .
No, and nobody has one. No official dMAT paper has been released, so anything advertised as a past paper is either invented or taken from a different exam. These questions are written to the official task formats — the same structures, constraints and answer types — which is the only honest way to rehearse for it.
Yes, a free one. It takes a moment and no card is involved. The account is what lets your score and per-subtest breakdown still be there when you come back, so you can measure a second attempt against the first rather than starting from nothing.
Thirty minutes, timed, for thirty questions. The clock is part of the point — most marks lost on a first attempt go to the format and the pacing rather than to weak reasoning.
Your score, a breakdown of accuracy by subtest so you can see which one is costing you the most, and a worked explanation for every question you attempted — including why the wrong options are wrong. Anything you skipped is listed separately, so you can go back and try it first.
No. The real examination runs about three and a half hours in total, including a break between the modules: a Core Module of three subtests of roughly twenty tasks in about twenty-five minutes each, and a Subject Module of about ninety minutes. This mock is a thirty-minute cross-section of all four task types, built so you can sit it in one go and still have time to review it properly.
The attempt submits itself and you land on your results. Anything you never reached is marked unanswered, which is deliberate: questions you did not get to are a pacing problem, and separating them from the ones you got wrong is the single most useful cut in the whole report.
Better not to. The Core subtests are designed to be solved mentally, so practising with a pad beside you quietly inflates your result and trains a habit you cannot use on the day. Sit the mock on a screen, with a clear desk, and let the discomfort show you where your working memory runs out.
No, and no practice platform can. This report gives you raw accuracy, your time, and how you performed on each task type. The official certificate reports each module on a scale from 0 to 200 along with a percentile rank, and that standardisation cannot be reproduced without official data. Treat the mock score as a diagnostic, never as a prediction.
Yes. A second attempt draws on the same free question set, so it is a check on your pacing and method rather than a fresh measurement — useful for seeing whether a correction has stuck, less useful as a new score. For fresh untimed work on one task type at a time, use the free practice questions instead.
The first dMAT in India is on 26 September 2026. Registration opened on 29 June 2026 and closed on 15 September 2026, certificates become available on 12 October 2026, and the fee is €150. It is a digital, single-choice examination in English, sat at licensed g.a.s.t. test centres — Ahmedabad among them. Always confirm dates and fees against the official sources before you plan around them.
30 questions, 30 minutes, worked explanations on every one. Free account required.
The mock is a diagnostic, not a course. Its job is to tell you which of the four subtests is furthest behind — and once you know that, what helps is volume on that one subtest rather than another full sitting. The Full Prep Pack opens 200+ original questions with a worked explanation on every one, subtest-by-subtest drills, and the full timed mocks. One payment, lifetime access, no renewals.
See what is included and what it costs — or keep going with the free questions on practice questions by subtest.
We link out so you can verify everything yourself. We are not affiliated with any of these bodies.