As many as you like. Every IB Marker account has unlimited access to every practice question, mock exam and multiple-choice round. The only soft limit is the rolling daily AI marking allowance on written answers, and even then your answers are saved and marked once it recharges.

Four. Written practice questions, from one-mark answers up to extended responses, marked against the question's markscheme. Multiple choice, four options with an instant verdict. Spot the Mistake, where you find the deliberate errors hidden in a worked answer. And mock exams, which are timed papers built from written questions.

There is no fill-in-the-blank format. See the Learning guide for the solo modes and the Multiplayer guide for the head-to-head ones.

If you have a practice set with questions still unanswered, it becomes the first card on the Practice Hub. The card names the subject and subtopic of your next question, how many you have answered, and one line estimating the XP still on the table and roughly how long the rest will take.

Resume set drops you back into the session with every answer you already typed still saved. Next to it, New quick set builds a fresh set on your usual subject in one click. See The Practice Hub.

Practice Mode is untimed. The clock on the session bar counts up rather than down, you build the set yourself or let a study plan build it, and your results page fills in question by question as the marks land.

Mock exams are timed, paper-style papers with a limit you cannot pause. You sit the whole paper, submit it, and it comes back with a total and the same point-by-point feedback. See Mock exams.

Yes, in Practice Mode. The custom builder lets you pick a subject, toggle any number of its topics as chips, search free text, and filter by question length and difficulty, then choose the individual questions or take a random five. You can also build a set from nothing but your bookmarked questions.

Solo multiple choice has no topic picker: you choose a subject and a length, and questions are sampled from across that subject's bank.

Every question carries a difficulty percentile within its own subject, so the rating is relative to the rest of that subject's bank rather than an absolute scale. Practice Mode's Easy, Medium and Hard filters are bands of that percentile, and Any is the default - usually the right choice while you are still finding your level. Mock exams pick a percentile range for you from your recent results.

It is a percentile ranking against the other questions in that subject. A 75% difficulty means the question is harder than 75% of that subject's bank. It is not a prediction of how you will do on it, and a hard question is worth more XP than an easy one.

Mock exams use adaptive difficulty. The system reads your recent results in that subject, places you at one of four levels - Foundation, Intermediate, Proficient or Advanced - and draws questions from the percentile range that level uses. It switches itself on once you have a couple of exams behind you in a subject, and the level used is shown in your results.

The exact bands are documented in the app: click How it works next to the IB Mock Exams heading. Adaptive difficulty affects mock exams only, never the questions you pick yourself in Practice Mode.

We suggest 1-2 mock exams per week per subject. That gives the adaptive system enough to track you while leaving time to review the feedback and practice weak areas in between. Closer to the real exams, more frequent papers help you build stamina under timed conditions.

Your profile has two halves. The identity panel carries your avatar on its large plate with your equipped border, your name, your level, your current league and your XP bar. Below it sits your performance record: a board of cards covering results, habits, subject trajectory, university badges and social form.

Customize record lets you rearrange that board with one of three layouts, and set whether visitors see everything, highlights only, or nothing at all. You draw your picture and equip a border in Avatar Studio, and a theme sets your colours - sitewide by default, or on your profile pages only. The standard badge showcase is five slots, chosen in Settings.

Everything competitive starts at the Multiplayer Hub in your sidebar. There is no matchmaking queue and no separate Join Game screen. Four modes sit on the hub - MCQ Battle, Spot the Mistake, Essay Duel and Custom Lobby - and the first three are ranked on one shared rating ladder.

Open a mode's card and you get two ways in: Quick play, which puts you in an open public lobby, or Create lobby, a private room with a share code you hand out yourself. Either route lands you in the same lobby box on the hub. See The modes.

The lobby box shows the subject and a live roster of everyone in the room. Press I'm ready; once everyone is ready and there are at least two of you, a short countdown runs and the game page opens for everybody at the same moment. Battles seat up to eight players, and you can keep inviting friends or sharing the join code while the room fills.

Closing the lobby box does not give up your seat - an In lobby pill in the top bar takes you back to it, and you only leave by pressing Leave inside the lobby. You can hold a seat in one lobby at a time. See The lobby.

Play now at the top of the hub is the one-tap route into a game. It reuses the setup you played last - "your usual" - looks for the open public lobby in that subject that has been waiting longest, and opens a new public one if there is nothing to join. Change setup swaps the subject or the length and saves it for next time.

One mode is featured each week, marked with a "1.5x XP this week" chip, and playing it that week multiplies what you earn by 1.5. Nothing to claim or opt into. See Quick Play.

It depends on the mode. MCQ Battle has no per-question clock at all - you work at your own speed and the side rail shows how far the rest of the field has got. Spot the Mistake times each round, and a longer passage gets a longer clock.

In Essay Duel the writing window scales with the question's mark total, so a heavier question gives you proportionally more time; the rail shows the phase, the marks and the time left. The in-app mode explainers on the hub carry the live figures, along with the How ratings work modal on your rank card.

Yes, and you never have to ask. In MCQ Battle, Spot the Mistake and Essay Duel, if you are still on your own after a short wait, AI opponents take the empty seats and the match starts. The in-app explainer puts that wait at 90 seconds, and real players who arrive first take those places instead.

They are rating-matched to you, they have their own names and their own place on the ladder, and they are marked (bot) everywhere they appear. Games against them are rated, but they move your rating less than a match against a person. Teacher class games are never filled with bots. See AI opponents.

Start a party. Open the Party tab in the hub's right rail and press Start a party; from then on, everyone in the party is invited automatically to any game you start. A party holds up to eight and you can be in one at a time. The host invites from your mutual friends, and members can suggest someone for the host to approve.

For a one-off, every online friend in the Friends rail has a 1v1 button that opens a private lobby on your last setup and invites them straight into it. The Challenge buttons on the Friends page and in a direct-message thread do the same thing. See Parties, challenges and invites.

The three ranked modes feed one Skill Rating ladder, divided into seven tiers from Rookie up to Legend. Win and you take rating from the players you beat; beating a higher-rated player is worth more. Your first few ranked games swing by more than usual while the ladder works out your level.

Custom Lobby games and teacher class games never touch your rating. The starting rating, the settling-in period and every tier's band live in the How ratings work modal on your rank card, because those values are tuned from time to time. See Rating and tiers.

Open Leaderboards in the sidebar. Every board lives on one page as five tabs: Season XP, Lifetime EXP, Multiplayer (competitive rating and tier), Questions answered with your success rate, and AI Models - the platform's bots, on the same live ladder students play for.

A scope switch above every board flips between all players, friends and your school, and your own row is always pulled into view. Nothing resets weekly: the season is the reset unit, and lifetime EXP and questions answered only ever go up. See Leaderboards.

A season is a calendar month, and your league is the race inside it. Season XP places you in one of seven leagues, from Bronze at the start up to Grandmaster at the top, with no promotion match and no mid-month demotion. Season XP counts raw effort with no badge multipliers applied, so a brand new account competes on even terms.

When the month rolls over your season XP drops to zero and everyone starts back in Bronze. Your lifetime XP, your level, your badges, your borders and your themes are never touched, and your final standing is archived. The Practice Hub shows your league, the full ladder and how long is left. See Leagues & Seasons.

Yes. Open Friends in the sidebar, search a classmate by username and send a request - they accept it from their notification bell. Search does not work on email addresses. Each friend row has a Challenge button: it opens a private lobby on the setup you played last and invites them straight into it.

Wherever your name sits next to a friend, it can carry a live activity line. Friends and squadmates see the subject and the round you are on; everybody else sees only the kind of thing you are doing, never the subject. Settings > Privacy > Appear offline drops that line and puts your online dot out everywhere. See What friends see you doing.

A squad is a team of 2 to 12 students who share a message board and compete together. Open Squads in the sidebar to search the registry by name, or paste a squad's invite code to join one straight away. You can be in one squad at a time, and a squad is either open to anyone or invite-only.

Squad Showdown is the weekly contest. Squads are drawn against the nearest squad by rating on Monday, every question you answer that week contributes (up to a daily limit per member), and Sunday freezes the score. It is scored on average XP per member rather than the total, so a squad of three competes evenly with a squad of eight. See Squad Showdown.

Two controls, deliberately different in strength. Mute hides that person's study-room messages and nothing else. Block stops direct messages, friend requests and game invites in both directions, hides each of you from the other's presence lists, makes your profiles private to each other, and ends an existing friendship.

Mute from a message in the study room; block from the Block button in a direct thread or on the person's profile. Both are managed in the Blocked & muted panel at the bottom of the Friends page, and neither is ever announced to the other person. See Block & mute.

The bell in the sidebar footer carries a badge with everything waiting for you, and opens a panel with three tabs. Social holds friend requests, unread direct messages, and game and party invitations - a friend request can be accepted or declined without leaving the panel. Academic holds school announcements, class updates, pending assignments and exam results. Progress holds new AI coaching insights.

Two things are deliberately not in the bell: level-ups, which get their own pop-up at the end of an activity, and class invitations, since your teacher adds you and the bell simply tells you about it. See The notification bell.

University badges are achievements modeled after real universities from around the world - a curated set across several countries, including Oxford, Cambridge, Imperial, MIT, Harvard, Berkeley, ETH Zurich, NUS, Melbourne and Monash. Each university carries real admissions data for its popular undergraduate programs, and pinning a program as your target makes practice in its required subjects level that university's badge. See University Badges.

Every university card shows its real, current world ranking - an aggregate of the three major published rankings (QS, Times Higher Education, and ARWU / Shanghai Ranking), taken as the median of each source's latest edition. Click the ranking chip on any card to see the per-source breakdown with years and links. Universities outside the global top few hundred show a factual regional line instead, like "#8 in Australia - THE". There are no invented difficulty tiers.

Badges turn real admissions goals into structured learning goals. Pin a target program (say, Oxford Medicine) and your practice in that program's required subjects counts toward the university's badge automatically - each level asks for more questions at a higher best-attempt average. Rather than practicing randomly, the badge keeps your practice pointed at the subjects your actual goal demands.

Each university badge has 6 progression levels: Bronze, Silver, Gold, Platinum, Diamond, and Master (shown as Levels I to VI). Each level has two gates - a question count and a minimum best-attempt average, from 15 questions at 65% for Bronze up to 50 questions at 90% for Master. Every level-up grants an XP bonus, and each level you hold adds a passive boost to the XP that builds your level, up to a cap. The full table is in the Badges guide.

No. Without a pinned program, practice in any of a university's program subjects still counts toward its badge. Pinning simply focuses the badge on your actual goal's required subjects and puts the pursuit in your spotlight. If a program expects subjects you haven't selected yet, you'll get a friendly heads-up when pinning - it never blocks you, since practicing those subjects is exactly how you make progress.

Each university carries real, sourced admissions data for the undergraduate programs most popular with IB applicants - typical conditional offers, per-subject HL/SL grade requirements, deadlines, and admission rates. The panel compares each program's requirements against your working grades computed from your best practice attempts, and you can pin up to 3 target programs as goals. Working grades are estimates from practice, not predictions of your IB results - always check the official program page.

The Achievements tab is your trophy cabinet: one plate per university showing its badge emblem, your highest level there, and a ledger of every level-up with the average you held, the questions counted, the XP awarded, and the date. Reaching Level VI (Master) completes a university's badge - a permanent record of the climb.

Yes. The standard allowance is five slots on the "University badges" card of your performance record. Fill them in Settings, in the Badges section, by clicking the badges you have earned. Other students see them when they open your profile from the leaderboard, your friends list or a message thread.

A border is the ring drawn around your avatar, and the one you equip follows you everywhere your avatar appears. There are 25 of them across six rarity tiers, and almost all of them unlock from things you have actually done - answering questions, adding friends, practice streaks, levels, multiplayer wins - checked against your real record, with no shop and nothing to claim.

Your unlocks are re-evaluated when you open Avatar Studio, so that is where a new one shows up first. See Profile Borders.

We do not publish an agreement-with-examiners percentage, because we do not have a set of student answers marked by trained IB examiners to compare against. Any figure we quoted would be agreement with another AI model rather than with a human, and we would rather say nothing than say that.

What we can tell you is what the marker is given: the question, its full markscheme, the subject and the marks available - nothing about you - and an explicit instruction not to invent credit for points the markscheme does not contain. Treat a single mark as an estimate and a trend across several answers as the signal. While a mark is pending you can tick the markscheme yourself, and there is a report control under every piece of feedback if you disagree with it. See How reliable is it.

Every marked answer comes back with four parts: your mark out of the total with an IB-style level, one bar per criterion showing exactly where the marks went, three notes (a strength, a concrete next step, and either a weakness to develop or a stretch challenge if you scored full marks), and short phrases lifted verbatim from your own answer and highlighted.

Under that sits a quiet row where you can rate the marking or report a problem with it. It is far more written feedback per answer than a teacher marking a class set can realistically give. See What your feedback contains.

Mark yourself. Your results page loads immediately, and while a mark is pending that question's card shows the real markscheme split into tickable lines: tick what you got while you wait. A running total tells you what you have given yourself.

When the real mark lands, your self-score sits next to it on the same line and each criterion is annotated - you agreed, you gave this, or you missed this. Seeing where you and the marker disagreed is the fastest way to learn what a markscheme actually rewards. Your ticks are kept on that device only.

The same marker and the same shared marking guide handle every subject. What changes per question is its own markscheme, which is what your answer is actually judged against, plus a fixed list of criterion names and their marks decided when the question was written.

The guide covers the rules of IB markscheme notation - caps as ceilings, a slash meaning "or", binding Accept and Do not accept instructions - and adds a rule for extended-response questions, where evaluation marks are earned by weighing arguments rather than listing them. Maths and chemical notation are typeset properly, but the marker reads text only, so type your working out step by step.

AI marks are for learning, not final grades, so there is no formal remark process. Two things you can do instead. Use Try again on the question: your previous attempt and its feedback are sent with the new one, so the marking reads as a follow-up rather than a fresh opinion.

And use the report control under the feedback. Reporting the mark flags how your answer was handled; reporting the question flags a defect everyone sees - unclear wording, a wrong model answer, off-syllabus content - and those are reviewed and can pull a bad question out of circulation.

No. The marker reads text only, so diagrams, graphs and hand-drawn working are not assessed - describe them in words instead. Mathematical working needs to be typed out step by step, and non-standard notation can be misread. Practice the visual components with your teacher and use IB Marker for the written ones.

Every marking on the site runs against one shared marking guide that carries a version number, and each mark records which version produced it. Criterion names and their available marks are pinned per question, so the rubric does not shift between attempts or between students, and a total can never exceed the marks a question is worth.

It is not perfectly repeatable, though, and the app says so: the same answer marked twice will usually get the same mark, but on longer, higher-mark questions it can move by one. Changes to the marking guide are A/B tested against the previous wording on hundreds of stored answers before they ship.

Nothing is lost. The allowance is a rolling 24-hour cap rather than a midnight reset, and if you reach it mid-session a notice appears on your results page saying how many answers are waiting and when it recharges. Those answers are saved and marked as soon as capacity returns.

You can keep practicing in the meantime, and each waiting card still shows the markscheme so you can self-check now and compare later. Multiple choice is not affected - the allowance only covers AI marking of written answers. See Your daily marking allowance.

Nothing. IB Marker is free for every student, every teacher and every school. There are no paid plans, no trial period that runs out, no subscriptions and nothing to pay - unlimited practice questions, mock exams, multiple choice, multiplayer, badges, leaderboards and the Insights page are all included.

The one limit is the daily AI marking allowance, a rolling 24-hour cap that stops any single account draining shared AI capacity. Reach it and your answers are still saved, and still marked once it recharges.

No. There is no premium tier, no in-app purchase and no payment details anywhere in the product. Borders, themes, avatars and badges are earned from what you have actually done, never bought. If a page ever asks you to pay for IB Marker, it is not us.

Ask your IB coordinator or IT administrator. School accounts are free too - no per-student billing, no tiers, no invoices - so the only question is whether somebody has signed your school up yet. If they have, enter the code on the Join School page and you are linked to your school's classes, assignments and teacher feedback.

A school code is not required to sign up. You can register on your own, use everything that does not need a class context, and join a school later.

You can link a school code at signup, or any time afterwards from Settings > School. To switch schools, use the same page to leave your current one and join a new one; your progress, badges and bookmarks carry over either way, and nothing about joining or leaving costs anything.

School codes come from your IB coordinator or IT administrator. If they don't have one, point them at the School Admin Guide - signup is self-serve, same-day and free.

Your results page never makes you wait behind a spinner: it loads straight away, a card at the top counts how many answers are done, and each question fills in as its mark lands. A whole set or a mock exam can take a few minutes to come back completely.

You do not need to keep the page open - marks are stored, so you can come back to them. Meanwhile, tick the markscheme on the pending cards and compare later.

Click "Forgot Password" on the login page and enter your email address. You'll receive a one-time reset link within a few minutes - check your spam folder if you don't see it. The link is valid for 30 minutes and can only be used once; request a new one if it expires. If your account was created with Google or Microsoft sign-in and you've never set a password, the same flow will let you set one for the first time.

First, try refreshing the page or logging out and back in. Progress updates may take a few seconds to sync. If the issue persists, clear your browser cache. Your data is saved on our servers, so it won't be lost - it's likely a display sync issue.

IB Marker works best on modern browsers: Chrome, Firefox, Safari, and Edge (latest versions). We recommend keeping your browser updated for the best experience. Internet Explorer is not supported.

Yes. You can request deletion of your account and all associated data by contacting our support team at support@ibmarker.com from the email address on your account. We will verify the request and permanently remove your data.

Deletion is irreversible - all your progress, badges, XP, and practice history will be deleted. Deleting your account doesn't affect anybody else at your school. If your account was provisioned by your school, your school administrator can also action this on your behalf.