Everyday AI Podcast – An AI and ChatGPT Podcast
Everyday AI Podcast – An AI and ChatGPT Podcast

Ep 751: Hands on with Google’s Gemma 4: How to Use The Open Source Model Locally and Why It Matters

April 8, 2026

AI Summary

5 min read

Google’s new Gemma 4 open-source model family, released under the permissive Apache 2.0 license, lets you run frontier-level AI on a consumer laptop for free, with full commercial freedom and no cloud dependency. The 31-billion-parameter model punches far above its weight, competing with models 20 times its size and matching the performance of the best proprietary models from just 15 months ago. For businesses and individuals, this changes the cost, privacy, and accessibility equation for running AI locally.

Why a 31-billion-parameter model matters

Parameter count is a rough proxy for model capability, but Gemma 4 breaks the usual relationship between size and performance. The 31-billion-parameter dense model scores a 1452 on the Arena ELO leaderboard, putting it in the same tier as open-source models that are 300 to 400 billion parameters in size. It ranks third globally among all open models. For context, GPT-4o was reportedly around two trillion parameters. Gemma 4 delivers comparable performance at roughly one-sixtieth the size.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

What you'll learn

  • 1 (00:00) **Introduction: The Local AI Revolution** - Jordan introduces Gemma 4 as a free, open-source model that can run locally, rivaling frontier models from a year ago.
  • 2 (01:33) **Why Gemma 4 is a Game Changer** - Explains how a 31B parameter model competes with models 20x its size.
  • 3 (03:59) **The "Pound-for-Pound" Champion of Open Models** - Uses a boxing analogy to describe Gemma 4's performance-to-size ratio.
  • 4 (05:36) **The Apache 2.0 License and the Return of Personal Software** - Explains the commercial freedom of the new license and predicts a resurgence of desktop software.
  • 5 (07:43) **Gemma 4 Capabilities and Model Variants** - Details the model's features and the four different flavors available.
  • 6 (10:10) **Hardware Requirements: The MacBook Pro Test** - Explains what hardware is needed to run each model variant.
  • 7 (13:10) **Performance Benchmarks: Gemma 4 vs. 2024's Best** - Compares Gemma 4's ELO score and parameter count to other open-source models.

+ Full timestamped outline available in the app

Show Notes

Is Vibe Coding dying already? 

Or, is will it be as essential to the next decade of work as the browser was for the past 20 years? 

And how can your company balance the speed and innovation side of vibe coding without accidentally leaking data or building a product that breaks more often than it works? 

We'll break down the basics on this Start Here Series deep(ish) dive into Vibe Coding. 

The Vibe Coding Boom: Why Vibe Coding isn't Going Away and How it's Both Good and Bad -- An Everyday AI Chat with Jordan Wilson


Newsletter: Sign up for our free daily newsletter
More on this Episode: Episode Page
Join the discussion on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Website: YourEverydayAI.com
Email The Show: [email protected]
Connect with Jordan on LinkedIn

Topics Covered in This Episode:

  1. Google Gemma 4 Open Source Launch
  2. Gemma 4's Apache 2.0 Licensing Explained
  3. Gemma 4 Model Variants & Hardware Requirements
  4. Small Language Models vs. Large Model Performance
  5. Benchmarking Gemma 4 Against Top AI Models
  6. Local AI Model Deployment Benefits & Privacy
  7. Hands-on Guide: Running Gemma 4 Locally
  8. Live Performance Test: Coding, Reasoning & Logic
  9. Instruction Following and Creative Output Demo
  10. Future Impact: Open Source AI for Businesses


Timestamps:

00:00 Gemma 4 release and features

05:13 Free AI models with GEMMA 4

06:39 Gemma's groundbreaking AI performance

10:26 Running AI models on MacBooks

14:32 Comparing model size and performance

16:48 Local AI benefits and privacy

22:11 Comparing AI models hands-on

25:01 AI solves river crossing puzzle

27:13 Fun trick question example

32:26 Brainstorming creative marketing strategies

35:48 Uploading files for transcript analysis

38:16 Comparing AI models for tone and style

40:12 Running AI locally on your device


Keywords: 

Go

Everyday AI Podcast – An AI and ChatGPT Podcast

More from this podcast

Everyday AI Podcast – An AI and ChatGPT Podcast →