Google has announced its innovative Gemini 3 AI model.

Google Introduces New AI Model Gemini 3

Alphabet-owned Google has announced the updated version of the anticipated Gemini 3 artificial intelligence model. Company executives stated that this new model has achieved a "significant leap" in reasoning and coding capabilities.

Highlights of Gemini 3

Gemini 3 will be instantly available in Google Search and other core services. This model not only offers users the ability to answer questions with interactive graphics but also advances its capabilities in processing text, images, and various types of media.

Advanced Reasoning and Outputs

Google CEO Sundar Pichai emphasized that the enhancement of reasoning capabilities based on inputs, as well as the significant increase in AI's capabilities over the past two years, are noteworthy. Pichai stated, "AI has progressed from reading text and images to understanding the complexities of an entire room."

Competition in the AI Race

Gemini 3 was designed to compete with innovations showcased by rivals like OpenAI and Anthropic. Google plans to increase the value of its infrastructure investments through the integration of the new model with core products and developer tools.

Interactive Application Development with Gemini 3

The new model can develop visuals and applications with interactive components based on user requests. For example, when travel planning is requested, Gemini 3 can generate clickable visuals.

Antigravity: New Development Platform

Additionally, Google introduced the new development platform Antigravity. This system allows developers to code autonomous agents and bring their projects to life more quickly.

Josh Woodward pointed out that the model not only understands inputs but can also generate results in completely new forms.