Google Gemma

Gemma is a family of lightweight, state-of-the-art open models to assist developers and researchers in building AI responsibly, built from the same research and technology used to create the models that powers the Google Gemini chatbot (formerly Bard) and AI-related tools for Workspace (formerly Google Duet AI).

Gemma is available in two different sizes — Gemma 2B and Gemma 7B — which both come with pre-trained and instruction-tuned variants and are both lightweight enough to run directly on a developer’s laptop or desktop computer, with CPU or GPU — and Google Cloud Platform with GPU and TPU acceleration. They are text-to-text, decoder-only large language models, available in English, with open weights, pre-trained variants, and instruction-tuned variants.

The models area available via Kaggle, Hugging Face, Nvidia’s NeMo, and Google’s Vertex AI