GitHub Actions for Free 24/7 AI Automation

Description and use

Running a local Llama3.2 model on a MacBook enables free, private AI tasks integrated into GitHub Actions.

Problem

Cloud AI services cost money per token and expose data to third‑party servers.

Outcome

Achieved zero‑cost, private, and fast AI automation by running a local LLM, enabling use cases like RAG, voice assistants, and code review on every commit.

Workflow

• Install Ollama via Homebrew
• Pull the llama3.2 model with ollama pull llama3.2
• Start the local server with ollama serve
• Create a Python script that posts prompts to http://localhost:11434/api/generate
• Integrate the script into a GitHub Action workflow for continuous automation

Evidence

Every token costs money. Every API call adds up. And your data goes to their servers.

Source

Now you have a GPT-4 level model running on your MacBook. Free. Private. Fast.

Source

I built 45 tools using this stack. All free. All local. All open source.

Source