AI

Set up self-hosted Quill to put a conversational AI agent in front of your existing SQL database: run the Docker appliance, connect a database and a model, build the agent, and embed the chat widget.

Learn how to use withContextPrefix() in RavenDB 7.2.3 to inject document-level context into every embedding chunk, improving vector search precision for large documents.

Learn how to connect multiple RavenDB AI Agents into a multi-agent system, where specialized sub-agents handle focused tasks while a coordinator routes work between them.

Learn how to attach binary files such as PDFs and images to RavenDB AI Agent conversations, reducing token usage while keeping files accessible across the conversation lifecycle.

Set up Context7 MCP with OpenAI Codex in VS Code to give your AI direct access to RavenDB docs — faster, context-rich vibe coding with fewer interruptions.

Build an AI agent with RavenDB: create an Orders Manager in Studio, wire up Query and Action tools, and scope conversation data per user.

Run semantic searches over your documents with RavenDB 7.0 vector search: generate embeddings on the server with no external tools, and query them through static indexes or auto-indexes.

Feed PDFs, images, and logs straight into RavenDB's GenAI feature. Three worked examples: image vector search without an image model, attachment-aware support triage, and PDF resume screening.

Build text-to-image and image-to-image product search on top of RavenDB vector search, using CLIP embeddings stored in documents and the original images kept as attachments.

External

Read about GenAI: How We Make You Work Less on the RavenDB.net news section

Build a semantic search API with RavenDB, Python, and FastAPI. Learn manual and automatic embedding generation with OpenAI, and run vector similarity search in under 50 lines of application code.

Learn how to use RavenDB's GenAI feature to automatically summarize, prioritize, and enrich documents. Step-by-step walkthrough with code examples and screenshots.

Add AI-driven vector search to a Next.js app using RavenDB's embeddings generation AI Task and a single vector.search RQL query, almost like using a regular search API.

Connect an AI provider like OpenAI, create an embeddings generation task, and query your documents by meaning with RavenDB vector search.

Extend RavenDB's indexing process with machine learning in C#: train an ML.NET model to flag toxic comments during indexing, and use a pre-trained Catalyst model to detect each document's language, all through AdditionalAssemblies and AdditionalSources.