Gemma 4 will change the game of AI! From coding to video understanding… Google’s new model is packed with awesome features

• Gemma 4 is now available in four different variants and supports 256K token context windows.
• Advanced Reasoning, Agentic Workflow and Code Generation are three major features provided in this model.
• Video, image processing and structured JSON output will make AI agents more capable and effective.

Tech company By Google Last week introduced the Gemma 4 Artificial Intelligence (AI) model. Compared to previous models, Gemma 4 has many amazing features. The previous model Gemma 3 focused on text and visual reasoning capabilities. But now the tech company says that the new open-source model introduced by the company offers agentic capabilities and advanced reasoning. The new model will be available in 4 different sizes. The new Large Language Model (LLM) will be available on Google’s developer platform. It can be downloaded from third-party repositories and run locally.

The first glimpse of iPhone Fold is revealed! 7.8 inch screen and A20 Pro chipset won the hearts of users, Video Viral on social media

Google introduces Gemma 4

The tech giant has announced the Gemma 4 AI model in a blog post. Also, this post has shared detailed information about the new model. The new model is available in four different sizes and configurations, including Effective 2B (E2B), Effective 4B (E4B), 26B Mixture of Experts (MoE) and 31B Dens. Its context window has also been increased to 256K tokens, up from 128K tokens in Gemma 3. Additionally, he has been natively trained in more than 140 languages. (Photo Courtesy – YouTube)

Another major change has been made in the new model compared to the previous model. Gemma 4 is now available under the permissive Apache 2.0 license. This allows it to be used for both academic and professional purposes. LLM can be used directly through Google AI Studio and Vertex AI. It can be downloaded from the company’s Hugging Face, Cagle and Olama listings.

Samsung made a big decision! ‘He’ messaging app to shut down soon; What is the reason? Read in detail

Gemma 4 is equipped with three key features: support for advanced reasoning, agentic workflow, and code generation. The new model with advanced reasoning is also capable of multi-step planning and deep logic. According to the data obtained, this model performs well in math and following instructions. This model is also capable of providing function calling and structured JSON output. So that users will be able to make their AI agents more powerful using this model. Google claims that this LLM supports high-quality offline code. Another notable feature of this model is its ability to process videos and images natively.