Weaviate is an open-source vector database that combines vector similarity search with structured filtering and graph-based data modeling, enabling semantic search applications where objects are stored as vectors alongside their properties and relationships. Built in Go for performance, Weaviate provides a schema-based approach to vector storage where data objects have defined classes with typed properties, and vectors are automatically generated or manually provided for each object. Key features include: modular vectorization (pluggable vectorizer modules — text2vec-openai, text2vec-cohere, text2vec-huggingface, multi2vec-clip for images — automatically generating embeddings during ingestion without separate embedding pipeline management), hybrid search (combining BM25 keyword search with vector similarity search using a configurable alpha parameter to balance both signals), GraphQL and REST APIs (flexible querying with filtering, aggregation, and exploration capabilities), multi-tenancy (efficient isolation for applications serving multiple users with separate data), generative search (integrating LLMs directly into the query pipeline — retrieve relevant objects then generate answers using connected models like GPT-4), cross-references (linking objects across classes enabling graph-like traversal and contextual retrieval), and HNSW indexing (Hierarchical Navigable Small World graph for efficient approximate nearest neighbor search with configurable recall-speed tradeoffs). Weaviate supports multiple deployment modes: self-hosted (Docker, Kubernetes, bare metal), Weaviate Cloud Services (fully managed), and embedded (in-process for development). The schema-based approach differentiates Weaviate from simpler vector stores — objects are typed with validated properties, enabling structured queries alongside semantic search (e.g., find articles semantically similar to a query where publication_date > 2023 and category = "technology"). Weaviate is widely used for RAG applications, e-commerce product search, content recommendation, and knowledge management systems.