Caching is the fastest way to cut latency and cost without touching the origin. This guide covers practical caching patterns for ApiNest consumers: HTTP cache headers, CDN behavior, client caches, surrogate keys, and invalidation strategies that work in production.

Core concepts

Freshness: how long a cached response can be served without revalidation.
Validation: confirming freshness with ETag/If-None-Match orLast-Modified/If-Modified-Since.
Stale-while-revalidate: serve stale instantly while refreshing in the background.

HTTP headers used by ApiNest

Cache-Control: public, max-age=30, stale-while-revalidate=120 on non-sensitive GETs.
ETag for content hash based revalidation when payloads are stable.
Some endpoints are non-cacheable (AI image/text); cache at your boundary if acceptable.

Client cache recipe (JavaScript)

const cache = new Map();
async function cachedFetch(url, opts = {}, ttlMs = 30000){
  const now = Date.now();
  const c = cache.get(url);
  if (c && now - c.t < ttlMs) return c.v;
  const res = await fetch(url, opts);
  if (!res.ok) throw new Error((await res.json()).message);
  const data = await res.json();
  cache.set(url, { v: data, t: now });
  return data;
}

CDN cache: when and how

Place a CDN in front of your proxy/server and respect origin cache headers. For personalized data, prefer cache segmentation (vary by API key or user bucket) or move to client cache only.

Surrogate keys for group invalidation

Tag responses with a header like Surrogate-Key: category:jokes. When data changes, purge by that key to invalidate hundreds of objects at once.

Troubleshooting

High origin load: increase max-age, add stale-while-revalidate.
Users see stale for too long: lower TTL and use background refresh.
Random misses: ensure URLs are normalized and query params ordered.

FAQ

Should I cache AI endpoints?

Not generally, but you can cache prompts that are identical for demos. Be explicit about TTL.

Where to cache first?

Start with a tiny client cache for repeated GETs, then add CDN for public data.

API Caching Strategies on ApiNest: Latency, Cost, and Scale