weaviate/weaviate

Health v2 · maintenance only

Adoption not part of the score

16754 stars · 1377 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Weaviate is an open-source, cloud-native vector database written in Go that stores both objects and vector embeddings, combining vector similarity search with structured filtering, hybrid keyword search, and integrated RAG. It supports automatic vectorization via model provider integrations (OpenAI, Cohere, Hugging Face, etc.) or importing pre-computed embeddings, with multi-tenancy, replication, and RBAC for production deployments.

Use cases

• build a semantic search engine over documents
• store and query vector embeddings with metadata filtering
• build a RAG pipeline with retrieval and generative models
• power a recommendation engine with similarity search
• search images by visual similarity
• build a chatbot with persistent vector memory
• run hybrid keyword plus vector search
• self-host a scalable vector database on Kubernetes

When to choose

• you need a production-grade vector database with filtering, replication, and multi-tenancy
• you want integrated embedding and generative model providers in one query interface
• you need hybrid search combining BM25-style keyword and vector similarity
• you plan to scale to billions of vectors on Docker or Kubernetes

When to avoid

• you only need a lightweight in-process vector index for a small dataset
• your workload is purely relational/SQL with no similarity search
• you need a fully embedded database with no server component (embedded mode is experimental)
• you want to avoid telemetry or cloud service dependencies entirely

Facets

service · maturity stable

vector-database search-engine database rag llm-inference databases machine-learning artificial-intelligence developer-tools cloud self-hosted go cross-platform semantic-search vector-search hnsw hybrid-search embeddings multi-tenancy grpc recommender-systems similarity-search search retrieval-augmented-generation docker kubernetes

Member repositories

For agents

markdown · JSON · MCP: product_card(name="weaviate/weaviate")