FSRS: The Smartest Spaced Repetition

What is FSRS?

FSRS (Free Spaced Repetition Scheduler) is a modern spaced repetition algorithm that outperforms older systems like SM-2 and Anki's default scheduler. Developed and peer-reviewed by the spaced repetition research community, FSRS uses machine learning principles to model your memory and predict the optimal time to review each flashcard.

How FSRS Works

When you study a card, you rate your recall on a 4-level scale: Again, Hard, Good, or Easy. FSRS analyzes your entire rating history and computes a personalized memory model with 17 parameters. It then schedules the next review exactly when the probability of recall drops to your target retention rate — by default 90%. Unlike SM-2 which uses fixed formulas, FSRS adapts to your unique forgetting curve.

Why FSRS Beats SM-2

• Personalized parameters — FSRS learns your unique memory patterns using 17 parameters instead of using the one-size-fits-all SM-2 formula.
• Target retention — You choose how much you want to remember (e.g., 90%), and FSRS optimizes all scheduling to hit that target.
• Fewer reviews, same retention — Peer-reviewed research shows FSRS requires 20-30% fewer reviews than SM-2 for the same recall rate.
• Open source & peer-reviewed — Transparent, continuously improved by the spaced repetition research community. Published benchmarks validate every release.

The 4-Level Rating System

Every time you flip a flashcard, you rate your recall. FSRS uses these four levels to continuously tune your memory model:

Built for Real Learning

FSRS continuously adapts as you study. Cards you consistently get right appear less often. Difficult cards get extra attention. The result: you spend time where it matters most, and your long-term retention steadily climbs above 90%. No wasted reviews. No forgotten cards. Just efficient, lasting memory.