Gemma 4 Arrives on Docker Hub: Lightweight AI Models for Every Workload

Introduction

Docker Hub is rapidly evolving into a central hub for AI models, serving millions of developers with a curated selection ranging from lightweight edge models to high-performance LLMs—all packaged as OCI artifacts. Today marks a significant milestone with the arrival of Gemma 4, the latest generation of state-of-the-art open models from Google, built on the same technology powering Gemini.

Gemma 4 introduces three distinct architectures designed to scale from low-power efficiency to high-end server performance. By packaging these models as OCI artifacts, they behave just like containers: versioned, shareable, and instantly deployable without custom toolchains. Developers can pull ready-to-run models from Docker Hub, push their own, integrate with any OCI registry, and plug directly into existing CI/CD pipelines using familiar tooling for security, access control, and automation.

And this is only the beginning. Over the next few weeks, Gemma 4 support will land in Docker Model Runner, enabling you not only to discover models on Hub but also to run, manage, and deploy them directly from Docker Desktop with the same simplicity you expect from Docker.

Docker Hub’s Growing GenAI Catalog

Docker Hub’s GenAI catalog already features popular models such as IBM Granite, Llama, Mistral, Phi, and SolarLLM, alongside applications like JupyterHub and H2O.ai, plus essential tools for inference, optimization, and orchestration. The addition of Gemma 4 further enriches this ecosystem, giving developers more choices for building intelligent applications.

What Docker Brings to Gemma 4

Run Efficiently at the Edge

Smaller Gemma 4 variants are optimized for on-device performance. Docker enables consistent deployment across laptops, edge devices, and local environments, making it easy to run AI models wherever you need them.

Scale Performance with Ease

From sparse to dense architectures, you can run any model like a container, making it simple to scale across cloud or on-premises infrastructure. Docker abstracts away complexity, so you focus on your application logic rather than infrastructure management.

One Command to Get Started

Getting started with Gemma 4 takes just one command:

docker model pull gemma4

No proprietary download tools, no custom authentication flows—just the same pull, tag, push, and deploy workflow you already use. By bringing Gemma 4 to Docker Hub, you get powerful models with a familiar, production-ready workflow.

What’s New in Gemma 4?

Gemma 4 redefines what “small” models can do, with architectures optimized across multiple sizes and use cases:

• Small & Efficient (E2B, E4B): Built for on-device performance with high throughput and low memory use.
• Sparsely Activated (26B A4B): Mixture-of-Experts design delivers large-model quality with smaller-model speed.
• Flagship Dense (31B): High-performance model with a 256K context window for long-context reasoning.

Key capabilities include multimodal support (text, image, audio), advanced reasoning with “thinking” tokens, and strong coding plus function-calling abilities.

Technical Specifications

Below is a summary of the three Gemma 4 architectures and their primary characteristics:

Architecture	Parameters	Primary Use Case	Context Window
Efficient (E2B, E4B)	2B–4B	On-device, edge	Standard
Sparse (26B A4B)	26B (active 4B)	High quality, reasonable speed	Standard
Flagship Dense (31B)	31B	Long-context reasoning	256K tokens

All models support multimodal inputs and are optimized for deployment via Docker Hub. For the latest details, refer to the official Gemma 4 documentation.

Getting Started with Gemma 4 on Docker Hub

To begin using Gemma 4, follow these steps:

1. Ensure you have Docker Desktop installed (version 4.30 or later recommended).
2. Open your terminal and run docker model pull gemma4.
3. Explore the model using Docker Model Runner (coming soon) or integrate into your own workflows.
4. Deploy your application—whether on laptop, server, or cloud—using the same docker run commands you already know.

For more advanced usage, visit the What Docker Brings to Gemma 4 section above.

Conclusion

With Gemma 4 now available on Docker Hub, developers have immediate access to cutting-edge AI models that are easy to deploy, scale, and manage. The combination of Docker’s familiar tooling and Google’s advanced architectures opens up new possibilities for building intelligent applications—from edge devices to data centers. Stay tuned for Docker Model Runner integration, and start experimenting with Gemma 4 today.