Mnemosyne (software)

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
Mnemosyne 1.x logo.png
Developer(s) Peter Bienstman
Initial release February 8th, 2006
Stable release
2.6 / December 11, 2017; 6 months ago (2017-12-11)
Repository Edit this at Wikidata
Written in Python
Operating system Windows, Linux, macOS, Android
Size 3.4 MB
Type Accelerated Learning & Memory Software
License AGPL v3 (except sync client), LGPL v3 (sync client)

Mnemosyne (named for the Greek goddess of memory, Mnemosyne) is a line of spaced repetition software developed from 2003 until the present.


  • Spacing algorithm based on an early version of the SuperMemo algorithm, SM-2,[1] with some modifications that deal with early and late repetitions.[2]
  • Supports pictures, sound, video, HTML, Flash and LaTeX
  • Portable (can be installed on a USB stick)
  • Categorization of cards
  • Learning progress statistics
  • Stores learning data (represented as decks of cards that each have a question and an answer side) in ".mem" database files, which are interoperable with a number of other spaced repetition applications
  • Plugins and JavaScript support
  • Review cards on Android devices.
  • Synchronization between other machines

Software implementation[edit]

Mnemosyne is written in Python, which allows for its use on Microsoft Windows, Linux, and Mac OS X. A client program for review on Android devices is also available but needs to be synchronized by the desktop program. Users of the software usually make their own database of cards, although pre-made Mnemosyne databases are available, and it is possible to import SuperMemo collections and text files.

Each day, the software displays each card that is scheduled for repetition. The user then grades their recollection of the card's answer on a scale of 0-5. The software then schedules the next repetition of the card in accordance with the user's rating of that particular card and the database of cards as a whole. This produces an active, rather than passive, review process.


Mnemosyne collects data from volunteering users, and is a research project[clarification needed] on long-term memory.[3]

An August 2009 version of the dataset was made available via BitTorrent;[4] a January 2014 version is available for download.[5] Otherwise, the latest version is available from the author, Peter Bienstman, upon request.[6]

See also[edit]


  1. ^ SM-2 Optimization of learning, Master's Thesis, University of Technology in Poznan, 1990 and adapted for publishing as an independent article on the web. (P.A.Wozniak, May 10, 1998)
  2. ^ "Principles", The Mnemosyne Project, retrieved June 3rd, 2008
  3. ^
  4. ^ Announcement; torrent index
  5. ^
  6. ^

External links[edit]