Looking for an Anki alternative? What I wanted, and the open-source app I built

Anki is brilliant, and a lot of people bounce off it. Spaced repetition works: people preparing for the MCAT basically live on Anki for a reason. But the app can feel like a cockpit, and writing hundreds of cards by hand is where most people quit.

I went looking for an alternative, didn't quite find what I wanted, and ended up building flashCards.io, an open-source flashcard app that an AI can safely write cards into. Here's what I learned about the options, including when you should just stick with Anki.

What Anki gets right

• Spaced repetition that works. Cards you know come back less often. Cards you miss come back soon.
• It's free and open source on computers and Android. The iPhone app is paid, and supports its development.
• Huge shared decks and add-ons. Whatever you're studying, someone has probably made a deck.
• Your data is yours. Decks are files you can back up and move.

Why people look for an alternative

• It's overwhelming at first. Note types, card templates, deck options. Powerful, but a lot to learn before your first review.
• Four answer buttons. Again, Hard, Good, Easy. Lots of people freeze deciding between Hard and Good.
• Writing cards takes forever. This is the real one. Making good cards is slow, and most people stop there.
• It looks its age. That doesn't matter to everyone, but a calm screen does make daily reviews easier to start.

The main Anki alternatives

If you want open source specifically, the real choices are Anki, Mnemosyne and newer projects like mine. Open source matters more than it sounds for flashcards: you may keep a deck for years, and you don't want it trapped in an app that changes its pricing or shuts down.

AI flashcard tools: the good and the bad

There's now a wave of AI flashcard tools for students. Upload a PDF or lecture notes, get a deck back in seconds. That fixes the biggest problem with spaced repetition, and it brings new ones:

• Too many cards. An AI will happily make 300 cards from one chapter. You'll never review them all.
• Cards that test recognition, not recall. Long questions that give the answer away.
• Confidently wrong facts. Rare, but when you're studying for the MCAT, one wrong card is one too many.
• Your cards live in their app. Often with no proper export.

The fix isn't avoiding AI. It's staying in charge of what it writes. Ask for fewer, sharper cards. Review a batch before you study it. And keep the cards somewhere you own.

What I built instead: flashCards.io

I wanted two things: a calmer place to study, and a safe way to let an AI fill it for me.

A calmer review

• Today shows how many cards are waiting, and one button to start.
• One card fills the screen. Space flips it.
• Two answers, Missed it and Got it, and the schedule adapts. If you miss Anki's finer grades, switch them back on.

Letting an AI in, safely

• It practises first. A dry run shows exactly what would change, without writing anything.
• Sending the same cards twice does nothing. Duplicates are skipped.
• Every change an AI makes is logged, so you can always check what it did.
• Mistakes come back clearly, naming the exact field that's wrong, so the AI can fix it itself.

No lock-in

• No account, no tracking, no internet needed. Everything lives in one folder you can back up.
• It imports Anki decks and spreadsheets, so you don't start from zero.
• It runs on Mac, Windows and Linux, and the code is on GitHub for anyone to build on.

So, which should you use?

• Already happy with Anki? Stay. Nothing beats a habit that works.
• Want notes and cards together? Try RemNote.
• Studying casually, with friends or a class? Quizlet is easy to start.
• Want something calm and open source, and want your AI to write the cards? Give flashCards.io a go, and tell me what's missing.

Curious how it was designed? The flashCards.io case study has the screens. And if you're building your own tool, here's how I go from Figma to code.