Google has officially announced the launch of Gemini 2.0, its latest AI model designed for the “agentic era.” CEO Sundar Pichai describes the new version as a significant step forward, stating that while Gemini 1.0 was focused on organizing and understanding information, Gemini 2.0 is all about making that information far more useful.
For Google, “agents” are systems capable of performing tasks on your behalf, using reasoning, planning, and memory to get things done.
Key features of Gemini 2.0
The first model available, Gemini 2.0 Flash, is already outperforming its predecessor, Gemini 1.5 Pro, on crucial benchmarks such as code generation, factual accuracy, math, and reasoning—all while processing data at twice the speed. The new version also supports multimodal output, meaning it can generate images mixed with text, making for a more dynamic, conversational experience. Additionally, Gemini 2.0 Flash can handle multilingual audio, which developers can customize in terms of voice, language, and accent.
One of the most notable features is its ability to call native tools like Google Search for more accurate answers and execute code when needed. This marks a step toward more advanced, practical AI applications.
Availability and developer tools
An experimental version of Gemini 2.0 Flash is already accessible to developers via AI Studio and Vertex AI, with general availability set for January. Google is also rolling out a new Multimodal Live API, which will allow real-time audio and video streaming inputs, such as from cameras or screens. These additions signal an exciting future for developers looking to build sophisticated, AI-powered applications.
Gemini 2.0 in Google’s consumer products
For end users, Gemini 2.0 is set to enhance the Gemini assistant, making it even more helpful. Users of both Gemini and Gemini Advanced will be able to try a chat-optimized version of Gemini 2.0 Flash in the Gemini app starting this week. A mobile app version will be available soon, and the model will also appear in AI Overviews in Google Search. This will allow more complex topics and multi-step questions—such as advanced math or coding queries—to be answered more effectively. A broader rollout of this feature is expected early next year.
With Gemini 2.0, Google is set to revolutionize how its AI tools assist users and developers alike, marking a new chapter in the evolution of artificial intelligence.
