weaviate and weaviate-examples
The main repository provides the core vector database engine, while the examples repository serves as complementary educational resources demonstrating how to use that database across different scenarios and integrations.
About weaviate
weaviate/weaviate
Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of a cloud-native database.
Built in Go for millisecond-scale performance on billions of vectors, Weaviate integrates vectorization from major providers (OpenAI, Cohere, HuggingFace) at import time or accepts pre-computed embeddings. It unifies semantic search, BM25 keyword filtering, image search, and generative RAG/reranking in a single query interface, with production features including horizontal scaling, multi-tenancy, replication, and RBAC for enterprise deployments.
About weaviate-examples
weaviate/weaviate-examples
Weaviate vector database – examples
Demonstrates semantic search, multi-modal retrieval (text-to-image via CLIP), and specialized NLP tasks (NER, Q&A, Named Entity Recognition) across diverse vectorization modules including Transformers and image encoders. Examples span Docker Compose deployments, Python/Node.js clients, and GraphQL queries—showing integration with frameworks like Haystack and tools like Prometheus for monitoring. Covers both pre-vectorized data ingestion and custom embedding workflows using BERT, SBERT, and PyTorch BigGraph.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work