RediSearch: Supercharge Redis with Full-Text and Vector Search

What Is RediSearch?

RediSearch is a powerful Redis module that brings full-text search, vector similarity search, and secondary indexing capabilities directly into Redis. With over 6,100 stars on GitHub, it transforms Redis from a simple key-value store into a full-featured search and retrieval engine — perfect for AI applications that need fast, scalable search at the edge.

Instead of running a separate search service like Elasticsearch alongside Redis, RediSearch lets you index your existing Redis data and query it with a rich query language. It supports exact-phrase matching, fuzzy search, numeric filtering, geo-radius queries, aggregations, and vector similarity search — all running natively inside Redis with minimal overhead.

Key Features

Full-Text Search

RediSearch provides stemming, tokenization, stop-word lists, highlighting, and spelling correction out of the box. You can run complex full-text queries on any Redis hash or JSON document with sub-millisecond latency.

Vector Similarity Search

With the rise of LLMs and RAG pipelines, vector search has become essential. RediSearch supports vector indexes using FLAT and HNSW algorithms, allowing you to store and query embeddings directly in Redis. This makes it ideal for semantic search, recommendation systems, and AI agent memory.

Secondary Indexing

Index any field in your Redis hashes or JSON documents — numeric, tag, geo, text, or vector. RediSearch automatically keeps indexes in sync with your data, so queries are always fast without manual maintenance.

Quick Start with Docker

Getting started with RediSearch is straightforward using Docker:

docker run -d --name redisearch -p 6379:6379 redislabs/redisearch:latest

Once running, connect with redis-cli and start creating indexes:

FT.CREATE idx ON HASH PREFIX 1 doc: SCHEMA title TEXT WEIGHT 5.0 body TEXT url TEXT
FT.SEARCH idx "hello world" LIMIT 0 10
FT.CREATE vec_idx ON HASH PREFIX 1 vec: SCHEMA embedding VECTOR FLAT 6 DIM 384 DISTANCE_METRIC COSINE

Why Self-Host RediSearch?

Self-hosting RediSearch gives you full control over your search infrastructure. There are no API rate limits, no data leaves your infrastructure, and you pay nothing per query. It integrates seamlessly with your existing Redis deployment — just load the module and start indexing.

Conclusion

RediSearch is an essential tool for any Redis user building AI-powered applications. Whether you need full-text search for a content site, vector search for a RAG pipeline, or a combination of both, RediSearch delivers production-grade performance with zero additional infrastructure complexity.