Local Chatbot with Retrieval Augmented Generation (RAG)
Build a 100% local RAG with n8n, Ollama and Qdrant. This agent uses a semantic database (Qdrant) to answer questions about PDF files. Tutorial !thumbnail.png Click here to view the YouTube…
- Use case
- Internal Wiki
- Difficulty
- intermediate
- Author
- Thomas Janssen
- Updated
- Jun 7, 2026
Required custom nodes
- agent
- lmChatOllama
- memoryBufferWindow
- textSplitterRecursiveCharacterTextSplitter
- formTrigger
- documentDefaultDataLoader
- chatTrigger
- vectorStoreQdrant
Local Chatbot with Retrieval Augmented Generation (RAG)
A working n8n automation that runs against a local model via Ollama - 28,717 views on the n8n template library. Build a 100% local RAG with n8n, Ollama and Qdrant. This agent uses a semantic database (Qdrant) to answer questions about PDF files. Tutorial !thumbnail.png Click here to view the YouTube…
What it does
Build a 100% local RAG with n8n, Ollama and Qdrant. This agent uses a semantic database (Qdrant) to answer questions about PDF files. Tutorial !thumbnail.png Click here to view the YouTube… It chains 10 nodes, integrating agent, lmChatOllama, memoryBufferWindow, textSplitterRecursiveCharacterTextSplitter.
Requirements
- n8n (self-hosted, free) to run the workflow
- Ollama serving a local model
- A GPU with enough VRAM for your chosen model (see best model per GPU)
Import it
Open the workflow on the n8n template library and click Use workflow to import it into your self-hosted n8n, then point its model node at your local Ollama endpoint (http://localhost:11434).
Use it with
Workflow by Thomas Janssen on the n8n template library. We link to the original to import; credit and the workflow JSON belong to its author.