zoom_in Click to enlarge

Gemma

▲ ▼

language

description Gemma Overview

Gemma is a family of open-weight large language models developed by Google DeepMind and released in February 2024. The models are built using the same research and technology as the proprietary Gemini models, but are provided as smaller, openly available variants. The initial release included two sizes: a 2-billion-parameter model and a 7-billion-parameter model, both designed to run efficiently on consumer hardware. Gemma is intended for developers and researchers who require accessible, open models for natural language tasks.

help Gemma FAQ

What is Google's Gemma model?

Gemma is an open-weight language model family developed by Google DeepMind. It is designed to be lightweight and accessible for developers to build and train AI applications locally or on cloud infrastructure.

How is Gemma related to Google's Gemini models?

Gemma is derived from the same research, technology, and infrastructure used to create Google's more powerful Gemini models. Essentially, Google adapted its flagship closed-source models into a smaller, open-weight format for public use.

What sizes does the Gemma model come in?

When released, the Gemma model family was made available in 2-billion and 7-billion parameter variants. These sizes allow developers to choose the right balance of performance and computing cost for their specific hardware.

When was the Gemma model released to the public?

Google released the Gemma model family in early 2024. It was launched under a permissive open license to encourage developer adoption and academic research.

swap_horiz

Looking for Gemma alternatives? Compare top competitors ranked & scored

arrow_forward