Google Gemma

Google Gemma is a family of open-source large language models released by Google in 2024. The models are distributed under the Apache 2.0 license, permitting both research and commercial use without additional licensing restrictions. Gemma represents Google’s contribution to the growing ecosystem of publicly available foundation models.

Model Variants and Design

The Gemma family includes multiple model sizes optimized for different use cases and computational constraints. Smaller variants are designed for efficient inference on resource-limited hardware, while larger versions provide capabilities comparable to contemporary large language models from other organizations. This range of sizes allows developers and researchers to select models appropriate for their specific deployment requirements and performance needs.

Training and Capabilities

Gemma models are built on the same research and infrastructure underlying Google’s Gemini models but released as open-source alternatives. The models are trained on diverse textual data and designed to follow instructions effectively. Like other large language models, they can perform tasks including text generation, question answering, and summarization across multiple domains.

Source Notes