Developers
Rate limits & caching
How our edge cache works and what to expect when calling the public API in volume.
Caching strategy
Every endpoint runs on Vercel's Edge Runtime and is cached globally. We use the Cache-Control header with s-maxage and stale-while-revalidate:
- Player profiles — fresh 5 min, SWR 1 h
- Skins / heads — fresh 1 h, SWR 24 h
- Server status — fresh 30 s, SWR 2 min
- News — fresh 60 s, SWR 5 min
Soft rate limits
There's no per-IP token bucket today. The cache is your friend — if your request matches a recent one, it's essentially free. Excessive traffic against uncached keys may be rate-limited by Vercel and our upstreams (Mojang and Crafatar both have their own limits).
Don't proxy as your own API
These endpoints are for the YuuTiers website and small community use. If you're building a service that needs millions of requests, run your own Mojang proxy with proper caching — both for your users and to be a good citizen of the ecosystem.