PANews December 12 news, according to Google's official blog, Google has released the new generation artificial intelligence model Gemini 2.0. Gemini 2.0 supports multimodal input such as text, images, videos, and audio, and features native image generation, multilingual text-to-speech (TTS), and other multimodal output capabilities. Compared to Gemini 1.5 Pro, the model speed has increased to twice as fast, and it has optimized multimodal reasoning, complex instruction execution, and tool usage capabilities, supporting calls to Google Search, code execution, and third-party functions.
The experimental version Gemini 2.0 Flash is now open to developers, and in January 2025, multimodal features will be fully rolled out, along with the launch of a multimodal real-time API to provide more application support for developers.