Repolode
Submit
kividbio

kividbio/kividb

A Redis- and Valkey-compatible in-memory key-value server written in Rust, with built-in full-text and vector search.

Open-source alternative to:RedisRedis Cloud

View on GitHubWebsite
Stars
272
Forks
6
Open issues
15
Last commit
2 days ago

What is it?

KiviDB is a server that speaks the Redis protocol (RESP2 and RESP3), so existing Redis clients, redis-cli and client libraries can connect without code changes. It stores data in a sharded in-memory key-value design that uses every core, and persists to snapshots (dump.kdb) or an optional append-only file.

Beyond the core commands, KiviDB implements Redis Stack-style FT.* commands for full-text (BM25), tag and numeric filtering, and vector nearest-neighbour search using FLAT or HNSW indexes on hash fields. Lua scripting and TLS are optional Cargo features, and a cluster mode with replication is available. It listens on port 6380 by default so it can run alongside a Redis instance on 6379.

The project is written in Rust, licensed under AGPL-3.0, and ships as a Docker image, release binaries and Linux packages. The maintainers publish benchmark results against Redis, Valkey-era alternatives and Dragonfly, measured on one controlled AWS setup.

Key features

  • 01Speaks the Redis protocol, so existing Redis clients work unchanged.
  • 02Multi-threaded sharded design that serves commands on every core.
  • 03Built-in FT.* full-text, tag, numeric and vector (FLAT/HNSW) search.
  • 04Snapshot persistence plus an optional append-only file.
  • 05Optional Lua scripting and TLS via build features.
  • 06Cluster mode and replication options.

Use cases

  • Replacing Redis or Valkey as a cache where higher throughput per server is needed.
  • Adding full-text and vector search to an application without running a separate search service.
  • Running a Redis-compatible store for local development or testing via Docker.
  • Serving vector similarity queries from hash fields with HNSW indexes.

Quick start

  1. 1

    Run the server with Docker

    docker run -d -p 6380:6380 kividb/kividb
  2. 2

    Check that it responds with PING

    redis-cli -p 6380 PING

Things to know

  • Protected mode is on by default, so remote clients need a password or bind address to connect.
  • Licensed under AGPL-3.0, which requires source disclosure for modified network services.
  • Version 1.1.0 and a relatively small project with 203 stars; benchmarks are vendor-published and from one controlled session.

Topics

#redis-compatible#key-value-store#in-memory#vector-search#full-text-search#database#rust#cache#dragonfly-alternative#drop-in-replacement

FAQ

What is kividb?
A Redis- and Valkey-compatible in-memory key-value server written in Rust, with built-in full-text and vector search. KiviDB is a server that speaks the Redis protocol (RESP2 and RESP3), so existing Redis clients, redis-cli and client libraries can connect without code changes. It stores data in a sharded in-memory key-value design that uses every core, and persists to snapshots (dump.kdb) or an optional append-only file.
Is kividb free to use? What license does it have?
kividb is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
Is kividb still maintained?
Actively maintained. The last commit to kividb was on Oct 18, 2026.
What can I use kividb for?
Replacing Redis or Valkey as a cache where higher throughput per server is needed. Adding full-text and vector search to an application without running a separate search service. Running a Redis-compatible store for local development or testing via Docker. Serving vector similarity queries from hash fields with HNSW indexes.
How do I install kividb?
Run the server with Docker: docker run -d -p 6380:6380 kividb/kividb Check that it responds with PING: redis-cli -p 6380 PING
Where can I get kividb?
The source code and installation instructions are on GitHub: https://github.com/kividbio/kividb

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects