🧠 Weaviate Flashcards
🔍 What is Weaviate?
Weaviate is an open-source vector database designed for semantic search and AI-native applications.
🧠 How does Weaviate support semantic search?
It enables vector-based retrieval using embeddings and allows combining it with traditional keyword filters.
📦 What types of data can it store?
Structured, unstructured, and vectorized data, including text, images, and metadata.
⚙️ Does Weaviate include vectorization?
Yes. Weaviate can integrate with models from OpenAI, Cohere, Hugging Face, or use its own modules to vectorize data.
🔧 What is hybrid search in Weaviate?
Hybrid search combines vector similarity with keyword-based filtering to improve retrieval accuracy.
📡 Does Weaviate offer REST or GraphQL?
Weaviate provides both REST and GraphQL APIs for flexible access and querying.
⚡ How fast is Weaviate?
It uses HNSW (Hierarchical Navigable Small World) indexing for high-speed approximate nearest neighbor search.
💼 Is Weaviate managed or self-hosted?
Both! You can deploy Weaviate in the cloud (Weaviate Cloud Service) or run it on-prem as open source.
🔐 What about authentication & authorization?
Weaviate supports API keys, OIDC, RBAC, and TLS for secure access and multi-user support.
🧪 Can Weaviate handle RAG workflows?
Yes, it’s widely used for Retrieval-Augmented Generation with LLMs to provide contextual results from vector storage.