Glossary · Performance

Vary Header

The Vary response header lists request headers that change the response, telling caches to store separate copies for each variation, such as Accept-Encoding.

Updated October 28, 2026

Vary is an HTTP response header that names the request headers a server used to choose its response. A cache must store and match separate versions based on the values of those headers.

Why it matters

Without Vary, a cache can serve the wrong variant. The classic case is compression: if one client gets gzip and another cannot handle it, a cache that ignores the difference may deliver compressed bytes to a client that cannot read them. Used badly, though, it fragments the cache and lowers the cache hit ratio.

Example

HTTP/1.1 200 OK
Content-Encoding: br
Vary: Accept-Encoding
Cache-Control: public, max-age=3600

The cache now keeps separate entries for Accept-Encoding: br, gzip, and none. Other legitimate uses are Vary: Accept-Language and Vary: Accept for content negotiation.

Vary: Cookie or Vary: User-Agent makes nearly every request unique. A CDN will then hold thousands of near-identical copies and rarely serve a hit. You can see what your site sends with the HTTP header checker.

Common mistakes

  • Varying on User-Agent or Cookie for pages that do not truly differ.
  • Omitting Vary: Accept-Encoding on compressed responses.
  • Using Vary: *, which effectively disables shared caching.

Get notified when this term updates

Plus occasional practical tips on uptime and monitoring, nothing else. Unsubscribe anytime.

Keep reading

Glossary · PerformanceCache Hit RatioCache hit ratio is the share of requests served from a cache rather than the origin server, a key measure of CDN and caching effectiveness.Glossary · PerformanceCache InvalidationCache invalidation is removing or refreshing cached content when the source data changes, so users stop receiving outdated copies from caches and CDNs.Glossary · PerformanceCDN (Content Delivery Network)A CDN is a geographically distributed network of servers that caches your content close to visitors, reducing latency and shielding your origin from load.Guide · PerformanceHTTP/2 and HTTP/3 Explained: What Changes and How to Enable ThemUnderstand multiplexing, QUIC and head-of-line blocking, when HTTP/2 and HTTP/3 actually help, and how to enable and verify them on nginx and Cloudflare.7 min readGuide · PerformanceImage Optimization Guide: WebP, AVIF, Sizing, srcset, Lazy LoadingChoose between JPEG, PNG, WebP and AVIF, size images correctly, use srcset and lazy loading, and batch convert with cwebp, avifenc and sharp.6 min readGuide · Hosting & CDNCDN Explained: What It Does, When You Need One, and How Caching WorksUnderstand how a content delivery network works, which headers control caching, what origin shield does, when a CDN helps or hurts, and how to verify it is working.7 min read

This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.