Oops I misunderstood it, thanks.
Btw is there a way to inflate Minutes today artificially 
This guy does 800 minute marathons (13 hour) Anki-time (not real time, which is understable) on a regular basis and not as a one-off which I would figure, is not humanly possible. Like is there an Anki option that controls how time is calculated
I mean set the max really high and casually do your reviews while strolling around and itās possible. Although, studying 14 hours/day isnāt something totally alien to me given I come from a city known for how hardcore students are.
1 Like
Yeah but Anki time though. 1 hour in Anki means 2 hours or so real time in practice. This means 14 hours of Anki non-stop. Full on focus. No breaks, no bathroom breaks. No eating. And probably no sleeping. Every day.
I guess this guy really does live inside Ankiā¦
Also Shige, your leaderboard doesnāt seem to register >10000 reviews for one day 
1 Like
Just want to say this is a great resource to really quantify this reaction speed discussion, thank you for sharing. 0.2-0.3 definetely seems like a reasonable cut-off, if a cut-off were to be implemented.
Yes, open a preset options, scroll down to Timers. If someone adjusts the time to an extreme and leaves the reviewer open when they step away, then it could be artificially elevated. But this would be hard to determine since some users may be reading large amount of text per card (even if itās not recommended to have such complex cards for most users).
I personally changed mine from default (60 seconds I think) to 120 seconds to more accurately reflect my time.
From the tool-tip in Anki:
Maximum answer seconds
The maximum number of seconds to record for a single review. If an answer exceeds this time (because you stepped away from the screen for example), the time taken will be recorded as the limit you have set.
2 Likes
Thanks for the report, I adjusted it.
Other than the way student says, if the user uses the add-on speed focus mode or auto advance the review can be done automatically. Also the leaderboards are open source and the decks are just numbers, so it is not technically difficult to generate AI code to cheat. (e.g. change the numbers to send, use a program or add-ons to generate dummy decks.)
It is possible if the user is able to study full time. e.g. It is not uncommon for people to work more than 15 hours every day at their jobs. Also a few percent of the population are short sleepers. They are rare but the leaderboard now has 8000+ users so there are likely to be more than 100+.
2 Likes
Personally I would put more time to this but I have school in backround. I can squeeze maybe extra 3 hours daily from brakes/travel/skip classes thanks to android app on phone. Weekdays can go 10+ hours if feel like it. Usually try to avoid more than that because then my brains are fried.
1 Like
Though not common because of the budget required, such devices and exercise equipment will make it possible to do 8+ hours or more of Anki with little or no fatigue. It is almost the same way to make remote work more efficient, and itās relatively easy because Ankiās review requires only the push of buttons.
- Remote, Gamepad
- Exercise ball
- Standing desk
- Exercise machine
- Treadmill, Exercise bike, Stepper
- Bring a tablet to the gym
- Home projector (review on Anki while lying on couch or bed)
Thatās a good point. I do 1ā2 hour daily walks on nature trails.
These days, everyone has a phone in their hands, so you donāt even get weird looks. Also with 15 seconds avg, my eyes arenāt constantly glued to the screen."
With, using anki in school, travel etc. Its very different versus sitting 10 hour infront computer !
1 Like
@Shigeyuki This is a perfect example of extremely sus behaviour. User āantijenā apparently at 100% retention achieved 8263 cards in just under 40 minutes!
3 Likes
Thats 0.3 sec avg.
In fps site max is around 0.2 but its very different because you only have to antipicate for color change.
But when have to process information max I think is around 0.3-0.5. Example flags, simple word to translate etc.
But thats maximum for 1 card only which is doable.
Real question is what is appropriate avarage for 1k cards. Personally I dont think can keep up even with 0.5 phase. With 1k+ questions there is no way to avoid brainfarts. Everyone will get stuck even with easy questions. This will raise avarage for 1+ sec eventually.
For retention doesnt matter much because hard button doesnt register some reason. Personally I dont use again button almost any.
I mark tougher questions with orange color and reset them daily. For rest I let anki algorithm do its thing.
1 Like
I think 100% retention would occur in a situation like these.
-
Users are using Hard instead of Again. According to a survey by Expertium about 10% of Anki users misuse Hard instead of Again. It is the same on the leaderboard so the retention rate of about 10% of users is always close to 100%. (Hard misuse occurs only in FSRS, it works fine in SM2, so if users are using SM2 they can use Hard instead of Again.)
-
Users are cramming and learning with filtered decks. (e.g.before exams) Also users can set their own button times for filtered decks, so it could be 100%.
-
User only 300 reviews in 31 days and no leaderboard history, so it looks to me that they are new to Anki. In that case it is possible that a beginner is reviewing a lot of cards incorrectly in the short term, and it is unlikely that they will be doing many reviews every day from now on.
-
Though it is unlikely, users can set the time to record a card to a minimum of one sec. In this case cards are recorded for a maximum of one sec and less than one sec so the time may appear shorter than the actual reviews.
-
Users may be reviewing as fast as possible to catch up on card delays. (e.g. there are 10,000+ overdue cards) Typically this is done using an add-on (e.g. FSRSHelper, Delay Overdue Cards), but if the user does not know of any add-ons they would review them manually.
Also if a user has cheats it is possible to intentionally increase the number of review sec, in other words the actual cheat can be made to take the most natural time (e.g. setting the auto advance to 5 to 10 sec). So I think users with very short times are more likely to be just fast reviewers or just using Anki in the wrong way.
just to let you know, sometimes when i look around on the leaderboard i get this. nothing really breaks but i get the error message lol
Anki 25.02.5 (29192d15) (ao)
Python 3.9.18 Qt 6.6.2 PyQt 6.6.1
Platform: Linux-6.8.0-60-generic-x86_64-with-glibc2.39
Traceback (most recent call last):
File "aqt.taskman", line 144, in _on_closures_pending
File "aqt.taskman", line 88, in <lambda>
File "aqt.operations", line 255, in wrapped_done
File "aqt.errors", line 44, in show_exception
File "aqt.utils", line 269, in showWarning
File "aqt.utils", line 304, in showInfo
TypeError: arguments did not match any overloaded call:
QMessageBox(parent: Optional[QWidget] = None): argument 1 has unexpected type 'RebuildLeaderbord'
QMessageBox(icon: QMessageBox.Icon, title: Optional[str], text: Optional[str], buttons: QMessageBox.StandardButton = QMessageBox.NoButton, parent: Optional[QWidget] = None, flags: Qt.WindowType = Qt.Dialog|Qt.MSWindowsFixedSizeDialogHint): argument 1 has unexpected type 'RebuildLeaderbord'
===Add-ons (active)===
(add-on provided name [Add-on folder, installed at, version, is config changed])
Anki Leaderboard - Compete with friends to boost motivation Custom by Shige ['175794613', 2025-05-15T00:23, 'None', mod]
Anki Simulator ['817108664', 2023-11-06T18:26, 'None', '']
Image Occlusion Enhanced ['1374772155', 2022-04-09T09:15, 'None', '']
Review Heatmap ['1771074083', 2022-06-30T03:43, 'None', '']
===IDs of active AnkiWeb add-ons===
1374772155 175794613 1771074083 817108664
===Add-ons (inactive)===
(add-on provided name [Add-on folder, installed at, version, is config changed])
Pokemanki Gold - Raising Pokemon with Anki Customized by Shige ['1677779223', 2025-06-07T21:42, 'None', mod]
````Preformatted text`
1 Like
Thanks for the error report, Iāll look into it.
2 Likes
offtopic but why my anki gives weird 1 sec lags especially when trying to create new decks etcā¦ I have 18k cards which aint really that much imo
not many addons. custom backround, review hotmap, leaderboard..
maybe like 400 pics total which aint much.
1 Like
The add-ons that are displayed in Ankiās home may take a little time to calculate, so try disabling them one by one to see which add-ons are causing the problem. Tools ā Add-ons ā Select Add-on ā Toggle Enable
Yeah custom backround was biggest cause. Suprisingly heavy.
1 Like
There are reports of similar problems in the comments of the Custom Background add-on page, so it seems that the recent update of Anki has caused something of a problem, perhaps you need to wait until the author updates the add-on.
1 Like