What Is FSRS? Free Spaced Repetition Scheduler, Explained

FSRS, short for Free Spaced Repetition Scheduler, is an open-source algorithm that estimates how likely you are to remember each flashcard and schedules the next review for the point where that probability falls to a target you choose (90% by default in Anki). It is not a new vocabulary learning trick. It is the calculation layer behind spaced repetition. Traditional schedulers move cards by fixed rules; FSRS schedules each card by your own recall history.

What is FSRS?

FSRS stands for Free Spaced Repetition Scheduler. It is an open-source spaced repetition scheduling algorithm developed by researcher Jarrett Ye and published by the Open Spaced Repetition project on GitHub. Its core job is simple: based on your recall result each time you review a card, it estimates how strong that memory currently is and predicts when it may drop close to forgetting. It then schedules the next review near that point.

FSRS describes each card's memory with three numbers, known as the DSR model (FSRS algorithm notes):

• Retrievability (R): the probability that you can recall the card right now.
• Stability (S): how many days it takes for retrievability to fall to 90%.
• Difficulty (D): how hard the card is for you, on a scale from 1 to 10.

For a learner, FSRS feels like this: you review a vocabulary card, mark whether you remembered it or forgot it, and the system decides when that card should return. If you remembered it, the interval gets longer. If you forgot it, the interval gets shorter. But FSRS does not do this with only fixed rules. It uses a model that estimates memory decay, so each card's schedule can better match your actual forgetting rhythm.

The research behind FSRS

FSRS grew out of published research on memory modeling and scheduling at the vocabulary app MaiMemo:

• Junyao Ye, Jingyong Su and Yilong Cao, A Stochastic Shortest Path Algorithm for Optimizing Spaced Repetition Scheduling, Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022), pp. 4381–4390.
• Jingyong Su, Junyao Ye, Liqiang Nie, Yilong Cao and Yongyong Chen, Optimizing Spaced Repetition Schedule by Capturing the Dynamics of Memory, IEEE Transactions on Knowledge and Data Engineering 35(10), 2023.

The FSRS4Anki README lists both papers as the starting point for the algorithm. The code, the optimizer and the benchmark data are public on GitHub.

How is FSRS related to spaced repetition?

This is the easiest point to confuse.

Spaced repetition is a learning method. Its core idea is to avoid studying everything at once or rereading the same list every day. Instead, reviews are spread over time and happen before you forget. The method uses the forgetting curve so each review happens when it is most useful for strengthening memory.

FSRS is a scheduling tool. It answers the question: “On which day should this card return?” Spaced repetition tells you to review across different time points. FSRS calculates whether this specific card should return tomorrow, next week, or three weeks later.

The relationship is: spaced repetition is the direction, FSRS is the navigation system. You can use spaced repetition with many scheduling systems. FSRS is one of them.

Why does vocabulary learning need scheduling?

Vocabulary learning has a practical problem: there are many words, and every word has a different familiarity level.

If you are learning 300 words, some are remembered after one encounter, some are still difficult after ten reviews, and some felt familiar yesterday but uncertain today. Manually deciding which words to review each day is almost impossible to do efficiently without missing important items.

A scheduling system solves this. It tracks the state of each card and automatically decides when it should appear again. Difficult words return sooner; familiar words get longer intervals. You only need to complete the reviews assigned for the day instead of deciding the entire schedule yourself.

Without scheduling, vocabulary learning becomes like reading a 300-word list from top to bottom every day, without knowing which words are stable and which need urgent reinforcement.

Traditional spaced repetition vs FSRS

In short, traditional spaced repetition works like “correct means longer, wrong means shorter” with relatively fixed rules. FSRS works more like “estimate this card's current memory strength from its history and predict when it will be close to forgotten.” It is closer to your actual forgetting rhythm, but the calculation behind it is more complex.

For most users, the practical difference is that FSRS scheduling is less likely to bring back cards too early when they are already easy, or too late when they have already been forgotten.

How is FSRS related to the forgetting curve?

The forgetting curve describes a simple phenomenon: after learning something, memory does not stay at the same level. It naturally declines over time. Without review, much of it may disappear within days.

Spaced repetition uses this pattern by scheduling reviews before memory drops too far. Each review raises memory strength again and slows the next decline. In theory, if reviews happen at the right moments, fewer review sessions can maintain memory for longer periods.

FSRS tries to make the “right moment” more precise. Instead of applying the same rule to all cards, it estimates memory decay for each card independently. This helps reviews land closer to the point before forgetting, rather than too early or too late.

The forgetting curve is the phenomenon. Spaced repetition is the strategy. FSRS is a calculation tool that makes the strategy more precise.

What Tomaru's learning data shows

Tomaru schedules every review with FSRS. Words reviewed on the schedule it sets are about 80% remembered when tested the next day, and 80–90% remembered after a week to a month.

The same gap means very different things for different cards. A week is fine for a word you already know, but a new word you didn't know the first time comes back less than half the time if its first review waits a week; reviewed the next day, about 60% are still there. That spread is the reason FSRS sets intervals per card instead of using one calendar for every word.

What is FSRS in Anki?

FSRS first reached Anki in 2022 as FSRS4Anki, a custom scheduling script and add-on. Since Anki 23.10, released on October 31, 2023, FSRS is built into Anki itself, and the Anki manual's FSRS section lists support in AnkiMobile 23.10, AnkiWeb and AnkiDroid 2.17 and later.

Inside Anki, you turn FSRS on in the deck options, set a desired retention and run the optimizer, which fits the FSRS parameters to your own review history. The manual sets 90% as the default desired retention and warns that above 90% the workload increases very quickly, because higher retention means shorter intervals and more reviews per day.

That control has a learning cost. To use it well, users need to understand desired retention, what the parameters mean, how workload changes after enabling FSRS, and whether different presets need different settings. This is not a problem for experienced Anki users, but for beginners it adds another layer of settings on top of an already flexible tool.

This is not a flaw in Anki. It is Anki's design direction: give users control over details. For some learners, that control is exactly what they need. For others, it is more than they want to manage. We compare the two approaches in Is Anki good for vocabulary?

Is FSRS good for vocabulary learning?

Yes. FSRS is a strong fit for vocabulary learning because vocabulary has several traits that need good scheduling.

First, vocabulary volume is usually large. Common words in a language can number in the thousands, and exam vocabulary lists often range from hundreds to thousands of words. Manual scheduling does not scale.

Second, familiarity differs by word. Some words stick after one encounter. Some remain difficult after many reviews. FSRS tracks memory state at the card level, so difficult words can appear more often while familiar words get longer intervals.

Third, vocabulary memory needs long-term maintenance. Remembering a word today does not mean you will remember it three months later. FSRS can bring long-term vocabulary back at appropriate times to prevent total forgetting.

The conclusion is direct: if you need to remember many words over time, spaced review with FSRS scheduling is one of the more efficient approaches.

How does Tomaru use FSRS?

Tomaru integrates FSRS spaced repetition into the vocabulary learning flow, so users do not need to configure or enable anything manually.

Tomaru's organised vocabulary decks cover exam preparation, daily situations, and common vocabulary, for example the JLPT N5, TOPIK I and Common Korean 3000 Words decks. Learners can start directly in the app without first organizing decks or judging deck quality. During review, your recall result for each card is used by FSRS to estimate memory state, and the system automatically decides when that card should appear again.

The process is transparent to the user: you review words, mark whether you remembered or forgot them, and the system handles scheduling. You do not need to understand retention, tune parameters, or study Anki's settings first.

Tomaru's core vocabulary features are free forever, with no ads. Its AI card creation feature can turn articles, lyrics, conversations, or news into reviewable vocabulary cards, expanding your learning material.

FSRS is not magic

One point matters: FSRS solves review timing, not every language-learning problem.

Even with good scheduling, each review still needs real recall. You should try to remember the answer first, then check it, instead of immediately flipping the card. Active recall is what strengthens memory. FSRS only helps make that action happen at the right time.

Vocabulary learning is also only one part of language learning. Knowing a word's meaning does not mean you can use it in the right context. Example sentences help with context, grammar practice gives words structure, shadowing supports language feel, and bidirectional translation improves expression. Tomaru places these exercises after card review so learned words can be used, not just recognized.

FSRS helps you remember vocabulary more efficiently. But after remembering words, you still need to use them.

How should you choose?

Choose based on your goal.

Choose Anki if you:

• Want full control over FSRS parameters and retention settings
• Need custom card formats and deck structures
• Have your own material and want to turn it into decks
• Are willing to study tool settings for maximum flexibility

Choose Tomaru if you:

• Want to start vocabulary learning with FSRS without studying settings first
• Need ready-made, organised vocabulary decks
• Want scheduling to work automatically without manual tuning
• Want card review to connect with grammar practice, shadowing, and translation
• Want core vocabulary features that are free forever, with no ads

FAQ

What is FSRS?

FSRS, or Free Spaced Repetition Scheduler, is an open-source spaced repetition scheduling algorithm. It tracks each card's difficulty, stability and retrievability, predicts when your chance of recalling it will fall to a target (90% by default in Anki), and schedules the next review for that point. Compared with fixed-interval scheduling, FSRS reflects each person's actual memory performance on each card.

What does FSRS stand for?

FSRS stands for Free Spaced Repetition Scheduler. “Free” refers to the open-source license: the algorithm, the optimizer and the research code are published by the Open Spaced Repetition project on GitHub.

Is FSRS the same as spaced repetition?

No. FSRS is a scheduling tool for spaced repetition, but they are not the same thing. Spaced repetition is the learning method: review before forgetting so memory becomes more stable. FSRS is the algorithm that calculates when a specific card should appear again.

How is FSRS related to the forgetting curve?

The forgetting curve describes how memory naturally declines over time. Spaced repetition uses that pattern to schedule reviews before forgetting. FSRS goes further by estimating memory decay for each card, so review timing can land closer to the point before forgetting instead of following fixed rules.

Is FSRS better than regular spaced repetition?

FSRS can be more precise than fixed-rule spaced repetition because it adjusts based on your actual memory performance for each card. In Tomaru's learning data, a word you already know is still 80–90% remembered after a week to a month, while a new word left a week before its first review comes back less than half the time, a gap that a single fixed interval cannot cover. For users who just want to study, the key is whether review feels smooth and sustainable.

Does Anki have FSRS?

Yes. FSRS has been built into Anki since version 23.10 (October 2023); before that it was available as the FSRS4Anki custom scheduler from 2022. Users enable it in deck options, set desired retention and can run the optimizer on their own review history.

Is FSRS suitable for vocabulary learning?

Yes. Vocabulary learning involves many words, different familiarity levels, and long-term memory maintenance. FSRS tracks memory state card by card, letting difficult words appear more often and familiar words get longer intervals. This is more efficient than manual scheduling or fixed-cycle review.

Is there a vocabulary app with FSRS that does not require setup?

Yes. Tomaru has FSRS spaced repetition scheduling built in, and its core vocabulary features are free forever, with no ads. Users do not need to enable FSRS or tune parameters manually. They can start with organised vocabulary decks in the app, and the system automatically schedules the next review based on each recall result.

Does Tomaru use FSRS?

Yes. Tomaru uses FSRS to schedule vocabulary card reviews. Each time a learner reviews a card and marks memory state, FSRS updates that card's schedule. The process runs automatically, so users can focus on learning vocabulary instead of studying algorithm settings.

Source: Tomaru anonymous learning data (2026).