Google Unveils ‘Gemma 3’ – Check Out Its Features and Capabilities

Google has introduced Gemma 3, the latest version of its open-source AI model. This new model is the successor to the previous Gemma models, which were built using Google’s Gemini 2.0 technology.

What is Gemma 3?

Gemma 3 is a large language model (LLM) designed to understand and generate human-like text. Unlike older models, this version can analyze text, images, and short videos, making it useful for a wide range of applications. One of its biggest advantages is that it runs smoothly on a single GPU, making it easier for developers to use without needing expensive hardware.

Stay informed with the latest news. Follow DXB News Network on WhatsApp Channel  

Different Sizes and Features

Google has launched Gemma 3 in different versions: 1B, 4B, 12B, and 27B. The company calls these its “most advanced, portable, and responsibly developed” open models yet.

These models are designed to run quickly and efficiently on different devices, from smartphones and laptops to powerful workstations. This makes it possible for developers to build AI-powered applications with faster performance and lower hardware costs.

A Year of Success

Last month, Google celebrated Gemma’s first anniversary, marking over 100 million downloads worldwide. The company says that Gemma 3 is the best single-GPU AI model in the world. It claims that this model outperforms Meta’s Llama, DeepSeek, and even OpenAI’s models when running on a single GPU.

Google has also optimized Gemma 3 to work better on Nvidia’s GPUs and other AI hardware, ensuring faster and smoother processing.

Better Image Processing and Safety Features

One of the key upgrades in Gemma 3 is its vision encoder, which now supports high-resolution and non-square images. This allows it to analyze and process different types of images more effectively.

Google has also introduced ShieldGemma 2, a new safety tool that helps filter explicit, dangerous, or violent content from images. This feature ensures that AI-generated content remains safe and appropriate.

Supports Over 140 Languages

Gemma 3 is designed to work globally, with built-in support for over 35 languages and pre-trained models for 140+ languages. This makes it an excellent tool for building applications that analyze text and short videos in different languages.

The model also comes with a 128k-token context window, allowing it to process large amounts of information at once. This means it can handle complex tasks and understand more detailed inputs than before.