Google Gemma

Gemma is Google DeepMind's family of lightweight, open-source language models designed to run efficiently on consumer hardware. Available in 2B and 7B parameter sizes, Gemma models achieve strong performance on benchmarks relative to their size. CodeGemma specialises in code generation, RecurrentGemma uses alternative architecture for efficiency, and PaliGemma handles vision-language tasks. Gemma models are Apache 2.0 licensed for commercial use and available on HuggingFace, Kaggle, and Google AI Studio for research and product development.