Responsible by design
Gemma prioritizes our AI Principles in its design. In order to ensure the safety and reliability of Gemma’s pre-trained models, we used automated techniques to filter out personal information and other sensitive data from training sets. Furthermore, we employed extensive fine-tuning and reinforcement learning from human feedback (RLHF) to align our instruction-tuned models with responsible behaviors. Rigorous evaluations, including manual red-teaming, automated adversarial testing, and assessments of model capabilities for potentially harmful activities, were conducted to understand and mitigate the risk profile for Gemma models, as outlined in our Model Card.
In addition, we are releasing a new Responsible Generative AI Toolkit together with Gemma to aid developers and researchers in prioritizing the creation of safe and responsible AI applications. This toolkit includes:
- Safety classification: A new methodology for building robust safety classifiers with minimal examples.
- Debugging: A model debugging tool to investigate Gemma’s behavior and address potential issues.
- Guidance: Access to best practices for model builders based on Google’s experience in developing and deploying large language models.
Optimized across frameworks, tools, and hardware
Gemma models can be fine-tuned on your own data to accommodate specific application needs, such as summarization or retrieval-augmented generation (RAG). Gemma is compatible with a wide variety of tools and systems:
- Multi-framework tools: Support for inference and fine-tuning across multi-framework Keras 3.0, native PyTorch, JAX, and Hugging Face Transformers.
- Cross-device compatibility: Gemma models run on various device types, including laptop, desktop, IoT, mobile, and cloud, providing accessible AI capabilities.
- Cutting-edge hardware platforms: Our collaboration with NVIDIA has optimized Gemma for NVIDIA GPUs, ensuring high performance and integration with advanced technology across data centers, the cloud, and local RTX AI PCs.
- Optimized for Google Cloud: Vertex AI offers a comprehensive MLOps toolset with a range of tuning options and one-click deployment using built-in inference optimizations. Advanced customization is available with fully-managed Vertex AI tools or with self-managed GKE, including deployment to cost-efficient infrastructure across GPU, TPU, and CPU from either platform.
Free credits for research and development
Gemma is designed for the open community of developers and researchers driving AI innovation. Free access to Gemma is available on Kaggle, with a free tier for Colab notebooks, and $300 in credits for first-time Google Cloud users. Researchers can also apply for Google Cloud credits of up to $500,000 to expedite their projects.
Getting started
For more information about Gemma and access to quickstart guides, visit ai.google.dev/gemma.
As we continue to expand the Gemma model family, we anticipate introducing new variations for diverse applications. Please stay tuned for upcoming events and opportunities to connect, learn, and build with Gemma.
We are excited to witness what you create!
“`
