• AI UNPLUGGED
  • Posts
  • Issue 31: Grok, Ollama & the Art of Inference

Issue 31: Grok, Ollama & the Art of Inference

Running AI Models Locally & Free

Author

Mike Murphy
November 07, 2025

🗞️ AI News of The Week: Grokipedia vs. Wikipedia

In Short: Elon thought Wikipedia went woke and got too biased, so he created his own unbiased version called Grokipedia. Of course, the Elon-haters say Grokipedia is biased, so maybe if you mix the two, you can find some truth.

Grokipedia is an AI-generated online encyclopedia developed by xAI that launched on October 27, 2025, as a rival to Wikipedia. Founder Elon Musk positioned it as an alternative to "purge out the propaganda" from Wikipedia, but early reception noted that many articles were copied from Wikipedia and promoted right-wing perspectives, conspiracy theories, and Musk's personal views.

Key Facts About Grokipedia

  • Type: AI-generated online encyclopedia.

  • Owner/Developer: xAI, founded by Elon Musk.

  • Content Creation: Entries are created and edited by the Grok large language model (LLM).

  • Launch Details: Launched on October 27, 2025, with over 800,000 articles.


🧠 AI Term of The Week: Inference

Inference is when an AI model (LLM) makes something from your input. After a model is trained (on tons of data), it doesn’t “think”…it infers. That means it uses what it learned to generate an output: a video, an image, a sentence, a sound. Every time you type a prompt or upload a photo and the tool gives you something back? That’s inference.

For creators, inference is where ideas turn into results.

💬 Quote of the Week:

❝


“To invent, you need a good imagination and a pile of junk"

– Thomas Edison

🧰 AI Tool of The Week: Ollama

🖥️ Run AI models locally on your own computer or VPS…no cloud, no subscriptions. Your data never touches the internet so it’s great for privacy and you can also use it on the beach, on the plane or wherever you are as no internet is required.

Ollama makes it insanely easy to run open-source language models like LLaMA, Mistral, and others with a simple terminal command. It’s fast, local, private, and lightweight and is perfect for developers, builders, or creators who want full control of their AI workflows and like to create or build stuff on the cheap (it’s 100% free!).

It’s one of my favorite AI Tools in my nerdy toolkit. I also like llamas. 🦙

👉 Try it at ollama.com

Turn natural language prompts into fully functional, deployable, AI-native applications in true prompt-to-product fashion.

👉 Check out Ollama


🛠 What I Made: How To Connect Ollama to n8n Workflows – No API Costs! (Part 2)

Description:
In this tutorial (Part 2) learn how to connect n8n to shared Docker Network and how to use Ollama AI Models in n8n automation workflows with no API Costs, Rate Limits and 100% free (excluding the cost of monthly Hostinger VPS Plan).

🤖 What You'll Learn

  • 👉 Inspect Docker Network

  • 👉 Terminal Tips

  • 👉 Edit n8n docker-compose.yml file

  • 👉 3-Step Nano Editor Process (Write out, Commit, Exit)

  • 👉 Stop & Restart n8n Docker container

  • 👉 Verify that n8n is connected to the shared network

  • 👉 Setup Ollama Chat Model Node credentials (internal URL)

  • 👉 Test Ollama Models

  • 👉 How To Pull, Remove & List Ollama Models

Get Hostinger VPS:
https://mikemurphy.co/vps


Mike Murphy — The AI Handyman 🧰
Helping creators & small businesses turn their content & documents into AI‑powered tools.