Gemma is a family of lightweight, state-of-the-art open models built from the same research and technology used to create the Gemini models. Developed by Google DeepMind and other teams across Google, Gemma models are designed to be customized, deployed anywhere, and integrated into your AI workflows.
🔗 Official Website | 📖 Documentation | 🤗 Hugging Face | kaggle Kaggle Models
New to Gemma? Explore our core guides, notebooks, and community resources:
| Repository | Description | Language |
|---|---|---|
| cookbook | A collection of practical guides, tutorials, and fine-tuning examples for Gemma open models. | Jupyter Notebook |
| docs | Source repository for official Gemma documentation. | Python |
| awesome-gemma | 😎 Curated list of awesome projects, libraries, tutorials, and tools built around the Gemma ecosystem. | Markdown |
Frameworks for extending Gemma with agentic skills:
- gemma-skills (
Python) — Reusable skills and patterns for Gemma model and agent interactions.
End-to-end sample applications demonstrating Gemma on the web, desktop, and edge devices:
- gemma-translator (
JavaScript) — Reference translation application powered by Gemma. - dinodesk-ai-companion (
Python) — Build a desktop AI companion robot powered by Gemma and Raspberry Pi.
We welcome contributions to our cookbooks, documentation, and awesome lists! Please check the CONTRIBUTING.md file within each individual repository for specific guidelines.
- Report Issues: Open an issue directly in the relevant repository.
- License: Code in these repositories is generally licensed under the Apache 2.0 License.
