On March 12, the leading tech giant Google introduced the Gemma 3 family of artificial intelligence (AI) models, which is an advanced open-source successor to the Gemma 2 series. The new model is capable of improved text and visual reasoning capabilities and supports more than 35 languages, with fine-tuning options extending to 140 languages.
Notably, Google claims these models are optimized to run efficiently on a single GPU or its proprietary Tensor Processing Unit (TPU). According to the blog post, Gemma 3 is a “collection of lightweight, state-of-the-art open models built from the same research and technology that powers our Gemini 2.0 models”.
Built With Gemini 2.0 Technology
According to Google’s official blog post, the Gemma 3 series uses the same core technology as the Gemini 2.0 models. It is designed as small language models (SLMs). They provide on-device performance while maintaining open-source accessibility.
The Gemma models have already gained traction, with over 100 million downloads and the creation of more than 60,000 model variants.
Google affirms that the Gemma 3 models outperform competitors, including Meta’s LIama-405B, DeepSeek-V3, and OpenAI’s o3-mini, based on LMArena leaderboard rankings.
The models are available in four sizes-1B, 4B, 12B, and 27B parameters- providing flexibility for developers working with diverse computational resources.
The Gemma 3 series introduces advanced text and visual reasoning features, which enable the analysis of images, text, and short videos. With a context window of 128,000 tokens and function-calling support, these models allow developers to build intelligent, agentic applications.
In the announcement, Google mentioned the safety measures implemented during development, stating that the models underwent rigorous risk assessments and find-tuning to align with internal safety policies.
Additionally, the company launched ShieldGemma 2, a 4B parameter image safety checker designed to prevent the generation of harmful, explicit, or violent content. The Gemma 3 models are available for download via Google’s Hugging Face listing and Kaggle.
Source: https://blog.google/technology/developers/gemma-3/
Latest Stories:
Samsung Unveils Galaxy Book5 Series with AI Features and Enhanced Battery Life
AI – WAQF 2025; Innovating Islamic Finance & Philanthropy
