Local LLM Chat Integration with n8n and Ollama

This workflow facilitates real-time interaction with self-hosted Large Language Models (LLMs) via n8n, using Ollama as the backend. It enables a seamless chat experience where user messages are received through a webhook, sent to the local Ollama server for processing, and the AI-generated responses are sent back to the user. The process involves a chat trigger node to capture incoming messages, a chain node to handle communication with Ollama, and the response is delivered immediately. This setup is ideal for developers and teams aiming to incorporate private, customizable AI chat capabilities into their workflows without relying on third-party cloud services.

Here’s a step-by-step overview:

1. The ‘When chat message received’ webhook captures user input.

2. The input is passed to the ‘Chat LLM Chain’ node that communicates with the Ollama LLM.

3. Ollama processes the prompt and returns a response.

4. The response is then sent back through the workflow, ready to be delivered to the chat interface.

Additional notes emphasize ensuring Ollama is installed and accessible locally, with configuration tips for Docker users.

This workflow is perfect for automation scenarios requiring private AI interactions, local AI tool integrations, or customized chatbots using self-hosted LLMs.