Do AI agents pick Memcached?
How often agents choose Memcached when a developer needs caching & kv — measured across Claude, GPT, Gemini.
95% CI 0%–3%
When developers ask an AI agent for caching, Memcached is picked 0% of the time — ranking #4 of 5 measured across Claude, GPT, Gemini.
1.3M package downloads / week
0 docs in FineWeb · 9 URLs in Common Crawl
4.5M installs / month ▲ 7% MoM
Get the full report
The per-model and per-surface breakdown for Memcached — where it wins, where it loses, and to whom — plus an alert when the Pick Rate moves.
About Memcached
A distributed in-memory key-value cache for speeding up dynamic applications.
Memcached is an in-memory caching system that stores arbitrary data (strings, objects) by key, reducing the need to repeatedly hit a database or slower data source. It supports single-server and multi-server cluster setups, using consistent hashing (compatible with libketama) to distribute keys across nodes so that adding or removing servers has minimal impact on key mapping. It's aimed at developers who need fast, scalable caching with configurable failure handling, connection pooling, and automatic failover.
Where to start
Stand up one or more Memcached server instances, then connect to them using the Node.js client by supplying a server address string, array, or weighted object along with any desired options such as pool size, retry behavior, and failover servers.