Google today announced Gemma 4 as its latest open model. It is “built from the same world-class research and technology as Gemini 3.”
Gemma 4 is available in four sizes for everything from Android devices to laptop GPUs and developer workstations/accelerators:
- 31B Dense: “#3 open model in the world on the industry-standard Arena AI text leaderboard”
- 26B Mixture of Experts (MoE): “#6”
- Effective 4B (E4B)
- Effective 2B (E2B)
For the 2 and 4 billion parameters, Google worked with the Pixel team, Qualcomm, and MediaTek. It can run on phones, Raspberry Pi, and Jetson Nano with “near-zero latency.”
Google says Gemma 4 goes from simple chat to “complex logic and agentic workflows,” with the company noting that it “outcompetes models 20x its size.” The edge models have a 128K context window, while the larger offerings go up to 256K, which lets you process long documents and repositories in one prompt.
All the models natively process video and images, which allows for OCR and chart understanding. There’s native audio input for speech recognition and understanding in the E2B and E4B models. Gemma 4 is natively trained on over 140 languages.
Other highlights include:
- Advanced Reasoning: Capable of multi-step planning and deep logic, Gemma 4 demonstrates significant improvements in math and instruction-following benchmarks that require it.
- Agentic Workflows: Native support for function-calling, structured JSON output, and native system instructions enables you to build autonomous agents that can interact with different tools and APIs and execute workflows reliably.
- Code Generation: Gemma 4 supports high-quality offline code, turning your workstation into a powerful, local-first AI code assistant.
Meanwhile, Gemma 4 is being released under the commercial/business-friendly Apache 2.0 license.
This open-source license provides a foundation for complete developer flexibility and digital sovereignty; granting you complete control over your data, infrastructure, and models. It allows you to build freely and deploy securely across any environment, whether on-premises or in the cloud.
You can access Gemma 4 in Google AI Studio (31B and 26B MoE) or in Google AI Edge Gallery (E4B and E2B).
Get the model weights from Hugging Face, Kaggle, or Ollama.
FTC: We use income earning auto affiliate links. More.
Comments