Google launched Gemma, a new suite of open-source AI language models, in partnership with DeepMind and other AI teams.

Gemma models are scaled at 2 billion and 7 billion parameters, aiming to surpass the performance of Meta's Llama 2 models.

The models are designed to operate on individual computers, allowing access via platforms like Kaggle and Hugging Face.

Google introduced a 'Responsible Generative AI Toolkit' alongside Gemma to ensure the safe and ethical development of AI technologies.

The initiative seeks to draw more developers to Google's Vertex AI cloud service and positions Google in direct competition with Meta's open-source AI efforts.