Design Manager
Gemma is designed with our Principles of AI in the first line. As part of making Gemma’s pre-trained models safe and reliable, we used automated techniques to filter out some personal information and other sensitive data from training sets. In addition, we used extensive refinement and reinforcement learning from human feedback (RLHF) to align our instruction-tuned models with responsible behaviors. To understand and reduce the risk profile for Gemma models, we conducted robust assessments, including manual red-teaming, automated adversary testing, and assessments of the model’s capabilities for risky activities. These assessments are described in our Model card.
We are also releasing a new one Responsible Generative AI Toolkit together with Gemma to help developers and researchers prioritize the creation of safe and responsible AI applications. The toolkit includes:
- Security Classification: We provide a new methodology for building robust security classifiers with minimal examples.
- Troubleshooting: A model debugging tool helps you investigate Gemma’s behavior and address potential issues.
- Directive: You can access best practices for model builders based on Google’s experience in developing and deploying large language models.
Optimized frameworks, tools and hardware
You can configure Gemma models with your own data to suit specific application needs, such as summarization or retrieval augmented generation (RAG). Gemma supports a wide variety of tools and systems:
- Multiframe Tools: Bring your favorite framework, with reference implementations for inference and refinement in multi-framework Keras 3.0, native PyTorch, JAX, and Hugging Face Transformers.
- Compatibility between devices: Gemma models run on popular device types including laptops, desktops, IoT, mobile and cloud, enabling widely accessible AI capabilities.
- Cutting Edge Hardware Platforms: We have worked with NVIDIA to optimize Gemma for NVIDIA GPUsfrom the cloud data center to the local RTX AI PCs, ensuring industry-leading performance and integration with cutting-edge technology.
- Optimized for Google Cloud: Vertex AI provides a broad set of MLOps tools 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 self-managed GKE, including deployment on cost-effective infrastructure across GPUs, TPUs and CPUs from any platform.
Free credits for research and development
Gemma is built for the open community of developers and researchers powering AI innovation. You can start working with Gemma today using free access to Kaggle, a free tier for Colab notebooks, and $300 in credits for first-time Google Cloud users. Researchers can also apply Google Cloud Credits up to $500,000 to accelerate their projects.
Getting started
You can explore more about Gemma and access quick start guides ai.google.dev/gemma.
As we continue to expand the Gemma model family, we look forward to introducing new variants for different applications. Stay tuned for events and opportunities in the coming weeks to connect, learn and build with Gemma.
We are excited to see what you create!