Generating embeddings
Create embeddings to enable AI-powered similarity search.
-
Embeddings are numeric vectors that represent the meaning, context, or relationships within the data (e.g., text or images). They allow you to search for similar content using Vector search, rather than exact matches.
-
RavenDB can generate embeddings automatically using Embeddings-generation tasks that you define. These tasks process text fields from documents and convert them into vectors using either an AI model from external providers (e.g., OpenAI) or RavenDB’s built-in model. The resulting vectors are stored in a dedicated collection in your database.
-
Alternatively, you can skip pre-generating embeddings via Embedding-generation tasks and let RavenDB generate them on-the-fly during dynamic vector search queries. Learn more in Vector search using a dynamic query.
-
RavenDB does not generate embeddings for non-text content such as images. However, you can import embeddings for such content from external providers and store them in your database for similarity search.
-
Embeddings also power other AI features in RavenDB, such as GenAI and AI agents,
which may use vector search queries to retrieve relevant documents in response to LLM requests.
Use cases
Embeddings-generation tasks can prepare your data for AI-powered search, analysis, and retrieval.
Common use cases include:
- Generate embeddings for thousands of documents, policies, and procedures to enable fast, semantic search.
- Process case law, contracts, and regulations to build a searchable legal knowledge base.
- Convert product descriptions, specifications, and reviews into embeddings for intelligent e-commerce search and recommendations.
- Transform blog posts, articles, and marketing materials into searchable vector embeddings.
- Generate embeddings for the semantic search step in RAG pipelines.
Technical documentation
Refer to the technical documentation to learn how to automatically vectorize data from your documents by configuring embedding generation tasks.
Learn more: In-depth articles
ExternalEmbeddings with RavenDB and External Models
Step-by-step setup guide for AI-powered semantic search, by Paweł Lachowski
ExternalThe integration of embeddings generation in RavenDB
The reasoning and architecture behind RavenDB's embeddings feature, by Oren Eini
