Features

Every feature, in detail

The front page shows these one at a time, which is the right amount to decide whether to keep reading and the wrong amount to decide whether to sign up. This is how each one actually works, in full, with the numbers it runs on — and where each one stops.

01 · THE MAP

A heatmap you cannot fake

Every subtopic of your course, coloured by what you proved.

A level is a running total of evidence, not a count of attempts. Every question carries a difficulty, that difficulty is banded into five heat levels, and the heat decides what a correct answer is worth — half a point for recall and one step, a point and a half at the hardest end of the paper. Eight points is mastery: below 3.5 you are Weak, 3.5 to 6 Developing, 6 to 8 Proficient, 8 and above Mastered. A subtopic with only a few questions asks for less, so every one can be mastered.

Which means an easy run cannot reach the top. Ten of the simplest questions is five points, and five points is Developing. The only route into the highest band is questions that were genuinely hard for you, which is the whole reason the engine prices by heat instead of counting right answers.

Wrong answers count too. Points get you to a level, and your last twelve answers in that subtopic have to hold it: once there are five or more, Mastered needs three in four of them right, Proficient 60%, Developing 40%, and below that it reads Weak. A subtopic you sat and got nothing right in is Weak, not grey. Every level also says how much it rests on, low confidence under five answers and high from twelve, so a Mastered built on three questions looks provisional rather than finished.

In short

  • 01

    Every subtopic of your course, from the official outline

  • 02

    Five levels, set by right answers and held by recent accuracy

  • 03

    Every level shows how many answers it rests on

  • 04

    No confidence slider. Anywhere. Ever.

Where it stops. A subtopic you have never been tested on is grey, not green. The map will not flatter you by assuming the parts you skipped are fine.

02 · DECAY

Topics that fade on their own

Prove something, leave it, and it quietly goes back on the list.

Day 0
Day 9
Day 16
Day 24

Fading is measured from your last correct answer, not from the last time the row was touched. That distinction matters: any quiz on a subtopic updates it, including one where every answer was wrong, so without the rule a fading topic could be cleared by sitting a quiz and failing it. Retention has to be re-proved, not revisited.

Memory fades on a curve, and the better you knew something the longer it lasts. A Mastered subtopic holds for four weeks without a correct answer, shows as Fading in weeks five and six, drops to Developing for week seven and then to Weak. Proficient holds three weeks and Developing two. One correct answer clears the fade, however long it has been.

In short

  • 01

    Mastered holds four weeks, then fades for two

  • 02

    Then Developing for a week, then Weak

  • 03

    One correct answer puts it back

Where it stops. It cannot know you revised something on paper. If you worked on a topic away from here, test it here and the fade clears.

03 · REVIEW

A mistake bank that schedules itself

Your own wrong answers, on a widening schedule.

Question you got wrongdue now
Question you got wrongin 1 day
Question you got wrongin 3 days
Question you got wrongin 5 days

Right each time and the gap widens until it drops out of the bank. Wrong and it starts over.

Every question you get wrong is kept against the subtopic it came from and returned to you on a widening gap: after a day, then three days, then five days. Answer it correctly each time and it leaves the bank for good. Get it wrong once and the schedule restarts from the beginning, because a question you have just missed is not one you nearly know.

The questions come back inside ordinary sessions rather than as a separate chore, so revision and review are the same act. The bank only ever holds questions you actually attempted — it is a record of the mistakes you made, not a model of the mistakes you might make.

In short

  • 01

    Every wrong answer returns, on a widening gap

  • 02

    Right each time and it leaves the bank for good

  • 03

    Wrong once and the schedule restarts

Where it stops. It only holds questions you have actually attempted. It is a record of your mistakes, not a prediction of which ones you would make.

04 · PAPERS

Timed papers with real pacing

Sit a paper the way your subject examines it, against a clock that tells you the truth.

42:081.4 / 1.6 marks per min

The tick is the pace you need. You are behind it.

You can assemble a paper from any combination of topics, or sit the structure your subject actually examines — Paper 1 and Paper 2, in their real shape. The clock runs against the pace the paper needs and shows marks per minute as you go, so you discover you are running slow during the paper, which is when it can still be fixed, rather than at the end.

Everything a paper produces feeds the same map as a single quiz, so sitting one is not a side activity that leaves your levels untouched. The limit is honest: a generated paper is built to the shape of a real one. It is not a past paper, and nothing here is scraped from one.

In short

  • 01

    Build from any topics, or sit Paper 1 and Paper 2

  • 02

    Live marks per minute against the pace you need

  • 03

    Results feed the same map as everything else

Where it stops. A generated paper is built to the shape of the real one. It is not a past paper, and nothing here is scraped from one.

05 · CALENDAR

Deadlines that change the plan

Put a test in and the plan reorders itself as it gets close.

12345678910111213141516171819202122232425262728

A test on the 11th pulls that subject up the plan from the 4th.

Put in your tests, your mocks and your IA deadlines, and the plan starts bending toward them. As a date gets closer that subject carries more weight in the ordering, so the evening tilts toward what is being examined soonest without you sitting down to re-plan it yourself.

A deadline changes the priority of a line but never hides the reasoning behind it — every item still states why it is on the list, so you can disagree with it. It does not read your school's calendar or your inbox. What is in it is what you put in it.

In short

  • 01

    Add your tests, mocks and IA deadlines

  • 02

    That subject rises in the plan as the date nears

  • 03

    Every item still says why it is there

Where it stops. It does not read your school calendar. What is in it is what you put in it.

06 · SESSIONS

A timer that follows you around

One number for whether tonight actually happened.

27:14

block 2 of 3 · keeps running as you move around

The timer starts once and follows you between pages, recording time against the subjects you actually worked on rather than the one you had open when you started. At the end of an evening there is a single number for whether that evening happened.

It measures effort and only effort. Time in a session never moves a subtopic up a level — only questions do that. This is deliberate, because hours logged is the easiest number in studying to inflate, and a tool that rewarded sitting still would be measuring the wrong thing.

In short

  • 01

    Starts once and follows you between pages

  • 02

    Recorded against the subjects you actually worked on

  • 03

    Measures effort. Never moves a level.

Where it stops. Time in a session is not evidence of anything. It never moves a subtopic up a level. Only questions do that.

07 · PREDICTION

A grade you can check the maths on

A running total built only from results, stated with its own confidence.

34 / 45you want38

Based on 41% of your syllabus, so it says medium confidence.

The predicted grade is a running total built only from quiz results. Nothing you tell it about yourself feeds in, because a prediction assembled partly from self-assessment is a prediction of your mood.

It also states how much of your syllabus it rests on, which is the part most predictions leave out. Three quizzes is labelled low confidence rather than dressed up as a forecast, and the number moves as the coverage grows. It measures the evidence you have produced so far. It is not a forecast of results day.

In short

  • 01

    A running total built only from quiz results

  • 02

    States how much of your syllabus it rests on

  • 03

    Three quizzes is labelled low confidence, not a forecast

Where it stops. It predicts from what you have been tested on here. It is a measurement of your evidence so far, not a forecast of results day.

08 · SYLLABI

A tutor that knows your syllabus

Questions, photos, PDFs and links, answered for your course.

Why is the derivative of ln x equal to 1/x?

S

Start from y = ln x, so x = e^y. Differentiate both sides… Now try one: d/dx of ln(3x)?

Photos, PDFs and links welcome. It coaches; it never writes your coursework.

Syllabi is a tutor that has your course in front of it: your subjects, your topics and the subtopic you are on. Ask it anything, send a photo of a question you are stuck on, a page of your notes or a link, and it explains the way your exam expects, then asks you something back to check it landed.

It is a coach, not a ghostwriter. It will not write your IA, essay or coursework, and it says so. Every request goes through daily limits and checks before it reaches the model, which is how a student project can offer it at all.

In short

  • 01

    Knows your subject, topic and subtopic

  • 02

    Reads photos, PDFs and web pages you send it

  • 03

    Explains, then checks you understood

Where it stops. It can be wrong, like any AI, and it never writes coursework for you. Premium only, with a daily allowance.

09 · MARKING

Drafts and answers, marked

Criterion by criterion, with your own sentences highlighted.

AResearch design4/6
BData analysis3/6
CConclusion5/6
DEvaluation2/6

14/24 · likely a 5 · each mark explained, with the next band spelled out.

Paste a draft of your IA, Extended Essay or TOK essay, or a Paper 1 or Paper 2 answer with its question, and Syllabi marks it against the published criteria for that piece: a mark per criterion, what earned it, and what the next band would need.

Your own sentences are highlighted where they help or hurt, and every marking is kept, so you can see a draft improve from one version to the next.

In short

  • 01

    IA, EE, TOK, and Paper 1 and 2 answers

  • 02

    A mark and a reason for every criterion

  • 03

    Every version kept, so you can see the change

Where it stops. An estimate from the criteria, not your teacher’s mark or the IB’s. Use it to improve a draft, not to predict a result.

10 · CARDS

Your notes, as flashcards

Photos, PDFs or pasted notes in; cards you can check out.

Enzymes — notes

· biological catalysts

· active site + substrate

· denature: heat, pH

→

What lowers activation energy?

An enzyme

Where does the substrate bind?

The active site

What does denaturing change?

The active site’s shape

Upload up to three photos of handwritten notes, PDFs or text files, or paste your notes, and Syllabi writes flashcards from exactly what is in them: definitions, facts, formulas, steps. Nothing it was not given.

The cards land in a list to check first. Edit any of them, untick the ones you do not want, then keep the rest. Your own cards are studied whenever you like, the ones you know least first.

In short

  • 01

    Photos of handwriting, PDFs and text files

  • 02

    Only what is in your notes, nothing invented

  • 03

    Checked and edited by you before they are kept

Where it stops. Blurry photos make thin cards. Pasting “term: definition” lines works for free without Syllabi.

11 · STREAKS

Streaks, goals and the whole picture

A reason to come back tomorrow, and a page that shows where the time went.

● 6-day streak · 2.5h this week

Time by subject, accuracy, when you focus best: all on one page.

Set a daily goal in minutes and keep a streak. Any study counts towards the streak, a quiz or the timer, and rest days on your plan never break it. The goal is what a good day looks like, not the price of keeping the streak.

The analytics page shows the rest: time by subject, a week to a year at a time, your accuracy over time and by subject, when in the week you study best, how long your focus sessions run, and what you have covered subtopic by subtopic.

In short

  • 01

    A daily goal and a streak you can see everywhere

  • 02

    Time by subject, accuracy, best study times

  • 03

    Week, month, three months, six months, a year

Where it stops. Time is counted from quizzes and the study timer. Reading a textbook with the timer off is invisible to it.

One subject is free for as long as you want it, and a school code opens the rest.

Start free with one subject