Gemma — Google DeepMind

Our most capable open models
A new level of intelligence for mobile and IoT devices
Frontier intelligence on personal computers
Our most advanced open models help developers create AI applications that run wherever users need them — from cloud servers to laptops and even phones.
Expanding beyond text to unify code, images, video, and audio in a shared embedding space
Built upon our Gemma 4 family and cutting-edge Gemini Diffusion research
Model compression for mobile and laptop efficiency
A unified, encoder-free multimodal model
Faster inference with multi-token prediction drafters
Byte for byte, the most capable open models. Purpose-built for advanced reasoning and agentic workflows
High-dimensional medical imaging interpretation in a compute-efficient model from Health AI Developer Foundations
A collection of translation models to help people communicate across 55 languages
Helping the AI safety community deepen understanding of complex language model behavior
Bringing bespoke function calling to the edge
The next generation of encoder-decoder models
The world's most capable differentially private LLM
The best-in-class open model for on-device embeddings
The compact model for hyper-efficient AI
A new collection of encoder-decoder Gemma models
Our most capable open models for health AI development
Frontier-level capabilities and mobile-first AI
A new level of intelligence for mobile and IoT devices
Frontier intelligence on personal computers
A new level of intelligence for mobile and IoT devices
Frontier intelligence on personal computers
A text diffusion model designed to maximize generation speed
Flexible encoder-decoder models that excel at deep contextual understanding
A collection of models trained for medical text and image comprehension
A modular classifier model for detecting policy-violating content and upholding safety standards
A text diffusion model designed to maximize generation speed
Flexible encoder-decoder models that excel at deep contextual understanding
A collection of models trained for medical text and image comprehension
A modular classifier model for detecting policy-violating content and upholding safety standards
From cutting-edge techniques to personalized models, discover how developers are using Gemma open models
Lentera runs an offline AI microserver for educators and students
Crane AI Labs improves Swahili language understanding in a lightweight model
Marine biologists and AI engineers partnered to develop DolphinGemma
Lentera runs an offline AI microserver for educators and students
Crane AI Labs improves Swahili language understanding in a lightweight model
Marine biologists and AI engineers partnered to develop DolphinGemma