早餐 Breakfast
早餐Breakfast
桜Cherry blossom
forêtForest
mercadoMarket
Capture · learn · go
The language app
that learns
with you.
Snap a photo, capture a conversation. Gravity finds the words inside the moments you're already living—and gently teaches them back to you.
You're in! One quick thing—check your inbox to confirm and you're all set.
Something went wrong — try again or email support@learngravity.com.
Coming soon to & 
From real life
桜 Cherry blossom
forêt Forest
mercado Market
swipe →
How Gravity works
Capture, select, learn.
A short loop, repeated whenever you encounter something you wished you understood.
- 01
Capture
Snap a photo of any text you want to read or remember, a menu, a sign, a manga panel.
- 02
Select
Tap any recognized words on the photo. Gravity gives you enriched vocabulary and grammar; you decide what to keep and Gravity does the rest.
- 03
Learn
Your words land in a personal deck. Gravity processes and determines the best words to surface for your daily lesson plan, understanding exactly where you're at and where you're going.
The method
How Gravity is built.
Four ideas from language and memory research that shaped the loop. Borrowed from people who studied it for a living, then translated into something you'd actually want to use.
-
Comprehensible Input
Linguist Stephen Krashen argued we acquire language by understanding messages slightly above our current level. He called it i + 1. Not isolated flashcards, not grammar lectures, just language we can almost understand, repeatedly.
Gravity surfaces exactly that. The words drawn from your photo are the ones the rest of the scene already tells you. You arrive at the meaning before you arrive at the rule.
-
Contextual Learning
Endel Tulving showed that what we remember is bound to where and how we encoded it. A vocabulary list memorized at a desk lives at that desk. A word learned at the market where you saw it lives at the market.
Every word in your Gravity deck carries its photo with it. When you remember the word, you remember the place. When you remember the place, you remember the word.
-
Dual Coding
Allan Paivio's theory: the brain stores verbal and visual information through two separate channels, and pairing them creates two retrieval paths to the same memory.
A photo and a word, learned together, are stickier than either alone. This isn't a feature of Gravity. It's the whole shape of it.
-
Free Spaced Repetition Scheduler
The forgetting curve isn't a hand-wave, it's measurable, and the right timing of review can flatten it. FSRS, the open algorithm by Jarrett Ye, models memory as three numbers per card (difficulty, stability, retrievability) and schedules reviews where they pay back the most. In practice it cuts roughly a third of the review load for the same retention as the older SM-2 schedulers.
Gravity reviews you on FSRS. You see each word right before you would have forgotten it, and not a minute sooner.
Work with Gravity
Bring Gravity to your group.
Whether you tutor a few students, teach a class, run a language program, or organize a meetup, we'd love to hear from you. Up for a demo? A coffee chat? A quick question? Pick whatever fits your week and we'll write back.
Email us
or copy hello@learngravity.com
Replies usually land within a day or two. Aio is slow with email but means well.
Frequently asked
Questions, answered.
The things we hear most often, in plain language. Missing one? Email hello@learngravity.com and we'll add it.
-
When does Gravity launch?
We're aiming to release on app stores before the end of May 2026—we're finalizing a few things and sending the app for review. Join the waiting list to be the first to know.
-
Which languages will be supported at launch?
We're scoping to Japanese at launch to make the early experience as polished as possible, with Chinese, Spanish, and French planned soon after. We also want to localize the app itself so non-English speakers can use Gravity too.
-
How is Gravity different from Anki or Duolingo?
Anki is powerful but you build every card yourself if you want personalized decks. Duolingo gives you an over-gamified curriculum. Gravity sits in between: it learns from the moments you're already living—a sign, a menu, a conversation—and turns them into a personal deck that reviews itself on the right schedule.
-
Do I need to take photos for it to work?
Not necessarily—you can also pull from your photo library (manga screenshots, a friend's texts, anything saved). Photos are the main input for now, but we're building toward direct text selection, document import, and audio capture. The idea is to start from real life, not a flashcard deck.
-
Will my captures stay private?
Yes. Your photos, notes, and recordings are yours—stored encrypted, never sold, and never used to train models outside of improving your own learning.
-
How much will it cost?
There's a generous free trial so you can build a real habit before paying. A subscription gets you 150–300 photo captures a month and a lot more down the line. Final pricing lands closer to launch, but expect less than your daily Starbucks order, per month.