Self-Hosted AI in 2026: Automating Your Linux Workflow with n8n and Ollama

Source: Dev.to

Why Combine Ollama and n8n?

• Zero latency – no external API round‑trips.
• Privacy – logs, secrets, and prompts never leave your hardware.
• No subscriptions – one‑time hardware cost, zero monthly fees.
• Full control – run any model you like (Llama 3.x, Mistral, DeepSeek, etc.).

Supported environment

• Any modern Linux distribution (Ubuntu 24.04+ or Debian 13 recommended).
• Ollama – the simplest way to run LLMs locally.
• n8n – “Zapier for self‑hosters” with built‑in AI nodes.
• Docker – for easy deployment and isolation.

Installing Ollama

curl -fsSL https://ollama.com/install.sh | sh

Verify the installation and pull a versatile model (e.g., Llama 3):

ollama pull llama3
ollama run llama3 "Hello, world!"

Deploying n8n with Docker Compose

Create a docker-compose.yml file:

version: '3.8'

services:
  n8n:
    image: n8nio/n8n:latest
    restart: always
    ports:
      - "5678:5678"
    environment:
      - N8N_HOST=localhost
      - N8N_PORT=5678
      - N8N_PROTOCOL=http
    volumes:
      - n8n_data:/home/node/.local/share/n8n
    # Allow n8n to reach Ollama on the host
    extra_hosts:
      - "host.docker.internal:host-gateway"

volumes:
  n8n_data:

Start the stack:

docker compose up -d

Open n8n at .

Adding an Ollama Node

1. Add an Ollama node to your workflow.
2. Configure credentials: set the URL to http://host.docker.internal:11434.
3. Select your model (e.g., llama3).
4. Connect the node to a trigger (HTTP request, Cron, etc.).

Example Workflow: Log Summarization Email

1. Node 1 – Execute Command

   tail -n 100 /var/log/syslog

2. Node 2 – Ollama

   Prompt: “Summarize these logs and highlight any security warnings or critical errors.”

3. Node 3 – Email / Discord

   Send the generated summary to your preferred channel.

Performance Tips

• GPU acceleration – Install nvidia-container-toolkit so Docker can use CUDA.
• Model quantization – 4‑bit or 6‑bit quantizations provide a good speed‑accuracy trade‑off.
• VRAM requirements
  • 7 B–8 B models: ~8 GB VRAM is sufficient.
  • 70 B models: 24 GB + VRAM (or a high‑end workstation such as Mac Studio).

Further Reading

• Ollama Official Documentation
• n8n Self‑Hosted AI Starter Kit
• Linux Automation Best Practices (2026)

Self‑hosting your AI isn’t just a technical choice; it’s a step toward reclaiming ownership of your tools. If you build something cool with this stack, feel free to share it in the comments.

Happy hacking!