The idea
Most anime recommenders fall into one of two camps: curated “top 100” lists that every new viewer already knows, or collaborative filters that keep recommending the same five massive franchises. Neither is great once you've watched a couple of seasons and know your own preferences. AnimeSwiper tries a third approach — a taste vector that updates with every interaction.
How recommendations work
Every anime title is embedded in a high-dimensional semantic space based on its tags, genres, and relationships. When you save, watch, or skip a title, your profile vector shifts a little toward (or away from) the things you liked. Discover then ranks fresh titles by vector similarity to your profile, blends in your explicit genre and tag signals, and drops in an occasional high-quality wildcard so your feed doesn't collapse into a narrow echo chamber.
What we do use
- Your swipes on Discover (save, watched, skip)
- Your explicit genre filters
- Aggregate AniList metadata (tags, genres, scores, popularity)
What we don't do
- We don't sell your data or your swipe history.
- We don't track you across the web.
- We don't boost titles because a studio paid for it — the only sponsored content is clearly labelled as “Sponsored”.
The data
Anime metadata is sourced from the excellent AniList public API and synced to our own database on a regular cadence, so Discover stays fast and works even when AniList is having a rough day. If you spot something wrong, AniList usually has the authoritative record — corrections there flow back to us on the next sync.
Monetization
AnimeSwiper is free to use. A small number of clearly-labelled “Sponsored” cards appear in Discover and on the Saved shelf via Google AdSense to cover hosting costs. See our Privacy Policy for details on how advertising cookies work.
Who's behind this
AnimeSwiper is built and maintained by a small team of anime fans who were tired of scrolling through the same “top 10 of all time” lists. The best way to reach us is email — details on the Contact page.