Automating Workflows with n8n and Local LLMs

Overview

n8n is a workflow automation platform that lets you connect apps and services using a visual node-based editor.

This playbook teaches you how to set up an AI-powered financial news summarizer that scrapes the AP News business section, extracts key headlines, and uses a local LLM running on your system to generate an investor-focused summary.

What You’ll Learn

• How to install and launch n8n
• Importing and configuring a pre-built workflow
• Connecting to Lemonade using the native n8n integration
• Understanding workflow nodes and data flow

What is Lemonade?

Lemonade is a local LLM serving platform built for AMD hardware. It provides an OpenAI-compatible API that runs entirely on your machine—your data never leaves your device.

In this playbook, we use Lemonade to serve a local LLM that n8n connects to for AI-powered tasks.

n8n includes a native Lemonade node (Lemonade Chat Model) that provides a first-class integration - no need for manual configuration. This makes connecting your local LLM to automation workflows straightforward.

Prerequisites

msiexec /i lemonade-server-minimal.msi /qn

sudo add-apt-repository ppa:lemonade-team/stable
sudo apt install lemonade-server

yay -S lemonade-server

lemonade --version

lemonade version x.y.z

node --version
npm --version

curl -o- https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh | bash

# Download and install Node.js:
brew install node@22

# Verify the Node.js version:
node -v # Should print "v22.22.1".

# Verify npm version:
npm -v # Should print "10.9.4".

msiexec /i lemonade-server-minimal.msi /qn

sudo add-apt-repository ppa:lemonade-team/stable
sudo apt install lemonade-server

yay -S lemonade-server

lemonade --version

lemonade version x.y.z

sudo apt update && sudo apt install -y podman docker-compose-plugin

sudo systemctl enable --now podman.socket

sudo podman run --rm docker.io/library/hello-world

Installing n8n

npm install -g n8n

podman compose up -d

n8n start

Launching Lemonade

Lemonade is the local server that will run a model and connect to n8n. Open a terminal and run:

lemonade run gpt-oss-120b-GGUF --llamacpp vulkan

lemonade run extra.gpt-oss-120b-GGUF --llamacpp vulkan

lemonade run gpt-oss-20b-GGUF --llamacpp vulkan

Alternatively, you can use the Lemonade GUI to choose and load a model. You can also experiment by changing to different backends, like rocm.

Setting Up the Workflow

Step 1: Sign Up or Log In to n8n

When you first open n8n, you’ll be prompted to create an account or log in:

1. Open http://localhost:5678 in your browser
2. Create a new local account with your email, or log in if you already have one
3. Once logged in, you’ll see the n8n dashboard

Step 2: Import the Workflow

We’ve provided a pre-built workflow that you can import directly:

1. Download the following workflow file: financial-news-workflow.json
2. Click Start from Scratch to open the workflow editor. Alternatively, click the + Button in the top left, and then Add workflow.
3. Click the … menu (three dots) in the top right bar and select Import from file
4. Select the downloaded financial-news-workflow.json file
5. The workflow will appear on the canvas

Step 3: Understanding the Workflow

The imported workflow contains 7 connected nodes:

Step 4: Configure Lemonade Credentials

Before running the workflow, you need to connect it to your local Lemonade server:

1. Double click the Lemonade Chat Model node in n8n
2. In the dropdown menu Credential to connect with select Create New Credential
3. Enter the values in the table below and click save.
4. Choose the relevant model that you have loaded in Lemonade Server.

Step 5: Test the Workflow

1. Ensure Lemonade is running with a model loaded
2. Click Execute workflow at the bottom center of the canvas
3. Watch each node execute from left to right—they turn green when complete
4. Click the AI Financial News Summarizer node to see the generated summary in the bottom pane.
5. Click the Convert to File node to download the corresponding text file in the bottom pane.

Understanding the AI Agent

The AI Financial News Summarizer uses a system prompt designed for financial analysis:

You are an AI financial analyst. Your role is to read, understand, and
summarize key financial news from today. The goal is to provide investors
with a clear and concise market overview to support better investment decisions.

Investor Outlook
Today's news points to [bullish/bearish/neutral] sentiment. Watch for
[economic event/earnings report] tomorrow, which could influence market direction.

The agent receives the cleaned news data and outputs a structured summary with market sentiment.

Saving Your Workflow

Click the workflow name at the top and rename it if desired. Workflows auto-save as you work.

Next Steps

• Schedule automation: Replace Manual Trigger with a Schedule Trigger to run daily
• Send notifications: Add a Discord, Slack, or Email node to receive summaries
• Try different models: Change the model in the Lemonade Chat Model node to experiment with different LLMs
• Customize extraction: Modify the HTML Extract node’s CSS selectors to target different news sections
• Try different backends: n8n also supports Ollama, LM Studio, and other local LLM backends

Explore n8n Templates

n8n has hundreds of pre-built workflow templates. Browse the official template library at:

https://n8n.io/workflows/

Search for “AI”, “LLM”, or “automation” to find workflows you can import and customize.

For more information, check out the n8n Documentation.