RAG pipeline diagram template
Retrieval-augmented generation in one picture. Documents are chunked and embedded into a vector store; the user’s question is embedded too; a retriever pulls the matching chunks and hands them to the LLM, which answers with citations.
Good for
- Explaining how a RAG system works to a team or a client
- Design docs for an LLM feature
- Posts and talks about retrieval-augmented generation
How to use it
- Press Use this template. It opens on the canvas in your browser, free, with no sign-up.
- Double-click any shape to rename it, or press / and type a line such as
Client -> API -> DBto add to it. - Share it as a link, present it with Shift+P, or export PNG, SVG or PDF.
In this diagram
- Retrieval-augmented generation
- Documents
- User question
- Chunk + embed
- Embed query
- Vector store
- Retriever
- top-k = 8
- LLM
- grounded prompt
- Answer with citations
- similar chunks
- vector
- context