# Weaviate — Open-Source Vector Database at Scale

> Weaviate is an open-source vector database for semantic search at scale. 15.9K+ GitHub stars. Hybrid search (vector + BM25), built-in RAG, reranking, multi-tenancy, and horizontal scaling. BSD 3-Claus

## Install

Save in your project root:

## Quick Use

```bash
# Docker (quickest)
docker run -d -p 8080:8080 -p 50051:50051 cr.weaviate.io/semitechnologies/weaviate:latest

# Or Weaviate Cloud (managed)
# Sign up at https://console.weaviate.cloud

# Python client
pip install weaviate-client

# Quick start
python -c "
import weaviate
client = weaviate.connect_to_local()
collection = client.collections.create('Articles')
collection.data.insert({'title': 'AI is here', 'body': 'Vector search makes it fast'})
results = collection.query.near_text(query='artificial intelligence', limit=2)
print(results)
client.close()
"
```

---

## Intro

Weaviate is an open-source vector database that stores objects and vectors, enabling semantic search at scale with sub-second query latency across billions of vectors. With 15,900+ GitHub stars and BSD 3-Clause license, Weaviate provides hybrid search combining vector similarity and keyword (BM25) search, built-in RAG and reranking, integrated vectorization from OpenAI/Cohere/HuggingFace, horizontal scaling, multi-tenancy with RBAC, and vector compression for cost efficiency.

**Best for**: Teams building production semantic search, RAG pipelines, or recommendation systems
**Works with**: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf
**Clients**: Python, JavaScript/TypeScript, Go, Java

---

## Key Features

- **Hybrid search**: Combine vector similarity and BM25 keyword search in one query
- **Built-in RAG**: Retrieval-augmented generation with integrated LLM support
- **Flexible vectorization**: Use OpenAI, Cohere, HuggingFace, or bring your own embeddings
- **Horizontal scaling**: Multi-node clusters with replication and sharding
- **Multi-tenancy**: Isolated tenant data with RBAC access control
- **Vector compression**: PQ and BQ compression for cost-efficient storage
- **Object TTL**: Automatic data expiration with time-to-live

---

### FAQ

**Q: What is Weaviate?**
A: Weaviate is an open-source vector database with 15.9K+ stars for semantic search at scale. It supports hybrid search, built-in RAG, reranking, and horizontal scaling across billions of vectors. BSD 3-Clause licensed.

**Q: How do I install Weaviate?**
A: Run `docker run -d -p 8080:8080 cr.weaviate.io/semitechnologies/weaviate:latest` for local setup, or use Weaviate Cloud for managed hosting. Python client: `pip install weaviate-client`.

---

## Source & Thanks

> Created by [Weaviate](https://github.com/weaviate). Licensed under BSD 3-Clause.
> [weaviate/weaviate](https://github.com/weaviate/weaviate) — 15,900+ GitHub stars

---
Source: https://tokrepo.com/en/workflows/492f7d14-9545-43b7-8f9c-626f895b912e
Author: AI Open Source