Google releases the first of its Gemini 2.0 AI models – NBC Los Angeles
Google released the first version of its Gemini 2.0 family of artificial intelligence models on Wednesday. Gemini 2.0 Flash, as the model is called, is available in a chat version for users globally while an experimental multimodal version of the model, with text-to-speech and image generation features, is available to developers.
Enhanced Features of Gemini 2.0
"If Gemini 1.0 was about organizing and understanding information, Gemini 2.0 is about making it much more useful," stated Google CEO Sundar Pichai. The latest large language model from Google outperforms its predecessors in areas such as code generation and providing factually correct responses from user requests. However, it falls short in evaluating longer contexts when compared to Gemini 1.5 Pro.
To access the chat-optimized version of the experimental Flash 2.0, Gemini users can select it from the model drop-down menu on desktop and mobile web. The multimodal version of Gemini Flash 2.0 will be available via Google's AI Studio and Vertex AI developer platforms.
Future Availability and Expansion
General availability of Gemini 2.0 Flash's multimodal version is expected in January, along with the introduction of more Gemini 2.0 model sizes. Google plans to expand Gemini 2.0 to more of its products in early 2025.
Competition in the AI Industry
Gemini 2.0 marks Google's continued efforts in the competitive AI landscape, facing competition from tech giants like Microsoft and Meta, as well as startups like OpenAI. Alongside the Flash model release, Google also introduced research prototypes aimed at developing more "agentic" AI models and experiences, allowing for advanced understanding of the world and proactive actions under user supervision.
During The New York Times' DealBook Summit, Pichai expressed interest in comparing Google's AI models against Microsoft's anytime, highlighting Google's commitment to AI advancement.
