Accountable by design
Gemma is designed with our AI Ideas on the forefront. As a part of making Gemma pre-trained fashions protected and dependable, we used automated strategies to filter out sure private data and different delicate information from coaching units. Moreover, we used in depth fine-tuning and reinforcement studying from human suggestions (RLHF) to align our instruction-tuned fashions with accountable behaviors. To know and scale back the danger profile for Gemma fashions, we carried out sturdy evaluations together with guide red-teaming, automated adversarial testing, and assessments of mannequin capabilities for harmful actions. These evaluations are outlined in our Mannequin Card.
We’re additionally releasing a brand new Accountable Generative AI Toolkit along with Gemma to assist builders and researchers prioritize constructing protected and accountable AI purposes. The toolkit consists of:
- Security classification: We offer a novel methodology for constructing sturdy security classifiers with minimal examples.
- Debugging: A mannequin debugging device helps you examine Gemma’s conduct and handle potential points.
- Steering: You possibly can entry finest practices for mannequin builders based mostly on Google’s expertise in creating and deploying giant language fashions.
Optimized throughout frameworks, instruments and {hardware}
You possibly can fine-tune Gemma fashions by yourself information to adapt to particular utility wants, corresponding to summarization or retrieval-augmented technology (RAG). Gemma helps all kinds of instruments and techniques:
- Multi-framework instruments: Carry your favourite framework, with reference implementations for inference and fine-tuning throughout multi-framework Keras 3.0, native PyTorch, JAX, and Hugging Face Transformers.
- Cross-device compatibility: Gemma fashions run throughout in style machine varieties, together with laptop computer, desktop, IoT, cell and cloud, enabling broadly accessible AI capabilities.
- Reducing-edge {hardware} platforms: We’ve partnered with NVIDIA to optimize Gemma for NVIDIA GPUs, from information middle to the cloud to native RTX AI PCs, making certain industry-leading efficiency and integration with cutting-edge know-how.
- Optimized for Google Cloud: Vertex AI offers a broad MLOps toolset with a variety of tuning choices and one-click deployment utilizing built-in inference optimizations. Superior customization is out there with fully-managed Vertex AI instruments or with self-managed GKE, together with deployment to cost-efficient infrastructure throughout GPU, TPU, and CPU from both platform.
Free credit for analysis and growth
Gemma is constructed for the open neighborhood of builders and researchers powering AI innovation. You can begin working with Gemma at present utilizing free entry in Kaggle, a free tier for Colab notebooks, and $300 in credit for first-time Google Cloud customers. Researchers may also apply for Google Cloud credit of as much as $500,000 to speed up their tasks.
Getting began
You possibly can discover extra about Gemma and entry quickstart guides on ai.google.dev/gemma.
As we proceed to increase the Gemma mannequin household, we sit up for introducing new variants for various purposes. Keep tuned for occasions and alternatives within the coming weeks to attach, study and construct with Gemma.
We’re excited to see what you create!