lux-db

lux-db / lux

Public

A Redis-compatable key-value store. 2-7x faster. Native vector support.

luxdb.dev database embeddings lux redis rust
185
10
100% credibility
Found Mar 19, 2026 at 185 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Lux is a high-performance, Redis-compatible in-memory data store with native vector similarity search, multi-threading for better speed, and automatic persistence.

How It Works

1
🔍 Discover Lux

You hear about Lux, a super-fast way to store and retrieve data that works just like your familiar tools but 2-7 times quicker with smart search built in.

2
💻 Get it running

Download the ready-to-go package and launch Lux on your computer or server with a single click – it's up and listening instantly.

3
🔗 Connect your apps

Point your existing programs or scripts to Lux's address; they connect seamlessly without any changes, feeling the speed boost right away.

4
✨ Add smart vector search

Store lists of numbers as vectors and search for the most similar ones super quickly – perfect for matching images, texts, or recommendations.

5
📈 Handle growing data

Load in tons of information; Lux spreads the work across your machine's power, staying snappy even under heavy use.

6
💾 Keep data safe

Set it to automatically save snapshots of your data at intervals, so nothing gets lost if something goes wrong.

🚀 Blazing fast results

Your apps now fly with lightning responses, reliable saves, and clever searches – enjoy the power without the hassle!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 185 to 185 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is lux?

Lux is a Rust-powered Redis-compatible key-value store that runs 2-7x faster by using all your CPU cores in a single process, dodging Redis's single-threaded limits. It supports 170+ Redis commands—strings, lists, hashes, sets, sorted sets, streams, pub/sub, Lua scripting, transactions—plus native vector storage and cosine similarity search via VSET/VSEARCH. Swap it into any Redis client like ioredis or redis-py with zero code changes, and get persistence snapshots plus BullMQ queue compatibility.

Why is it gaining traction?

On GitHub, lux-db/lux stands out in github lux ai searches for built-in HNSW vector indexing that skips sidecar databases, delivering sub-ms queries on 10k+ embeddings. Devs love the linear scaling on modern hardware—benchmarks crush Redis on SET/GET/LPOP at pipeline depths—without client-side sharding or cluster ops. MIT license and Docker-ready deploys make it a low-risk speed upgrade.

Who should use this?

Backend engineers scaling session caches or leaderboards past Redis's core limit. AI/ML teams building RAG or agent memory with Redis-protocol vectors, no Pgvector hassle. Node.js/Go/Python queue runners sticking with BullMQ but needing multi-core throughput.

Verdict

Grab lux for prod prototypes if Redis perf flatlines—solid tests (266+), reproducible benches, and TypeScript SDK shine—but at 185 stars and 1.0% credibility, treat it as alpha for high-availability yet. Watch for cluster mode.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.