Back to recipes
DeploymentIntermediate35 minVerified 29 days ago

Deploy n8n with Local LLM Tools

Self-host n8n workflow automation and connect it to local Ollama models for private, agent-driven automations.

n8nollamaself-hostingworkflow-automationlocal-llm
The promise

n8n is a workflow automation tool with native AI nodes. When self-hosted and paired with local LLMs, it becomes a private automation engine for agent-like tasks: classify emails, summarize documents, and route data without sending sensitive content to cloud APIs.

What you'll get
  • n8n running in Docker on your own server.
  • Ollama available to n8n workflows via the local network.
  • A sample workflow that calls a local model from n8n.
Prerequisites
  • Docker and Docker Compose installed.
  • Ollama installed and serving on http://localhost:11434.
  • A model pulled, such as qwen3.5:9b.
Troubleshooting
  • n8n cannot reach Ollama: Ensure Ollama binds to 0.0.0.0 or use the Docker host IP.
  • Model not found: Pull the model first with ollama pull qwen3.5:9b.
  • Slow responses: Local models need warm GPUs; cold inference is much slower.
Best fit

Teams that want no-code automation with local intelligence for privacy-sensitive or cost-sensitive workflows.

Steps

1
services:
  n8n:
    image: docker.n8n.io/n8nio/n8n
    restart: always
    ports:
      - "5678:5678"
    environment:
      - N8N_BASIC_AUTH_ACTIVE=true
      - N8N_BASIC_AUTH_USER=admin
      - N8N_BASIC_AUTH_PASSWORD=change-me
    volumes:
      - ~/.n8n:/home/node/.n8n

docker compose up -d
2

In n8n, go to Settings > Credentials and add an HTTP Request credential pointing to http://host.docker.internal:11434 (on Linux you may need http://<host-ip>:11434).

3
  1. Add a Schedule trigger.
  2. Add an HTTP Request node that POSTs to /api/generate with the model name and prompt.
  3. Add a Set node to parse the response.
  4. Save and test.

Example payload:

{
  "model": "qwen3.5:9b",
  "prompt": "Summarize the following text in one sentence: {{ $json.body }}"
}

Recipe verified 2026-07-01. Commands are tested but your environment may differ.

Browse related services