The end of 2024 is accompanied by a fairly spectacular race for technology, particularly in terms of artificial intelligence. Thus, OpenAI was able to launch Sora very recently, to which Google immediately responded with its own video generation model, Veo 2, which is part of the new generation of Gemini 2.0. And Google doesn’t stop there: at the same time, the firm is today launching Gemini 2.0 Flash Thinking, the objective of which is to step into the footsteps of OpenAI, with its o1 model.
Advertising, your content continues below
A model more efficient than all the others
Google’s desire to establish itself in the AI market is now palpable. With Gemini 2.0 Flash Thinking, the American giant’s goal is to position itself on the market for experimental models capable of offering advanced reasoning. This new model can “think out loud” and thus explain its reasoning for solving a problem to its user. It also benefits from a multimodal approach, capable of analyzing text, images and audio content, all at the same time. In doing so, Gemini 2.0 Flash Thinking turns out to be more impressive than ChatGPT-o1: the latter does not take into account audio files, in particular.
With this model, Google intends to offer more significant performance than that of already existing models. Thus, on the Chatbot Arena LLM ranking, Gemini 2.0 Flash Thinking quite simply takes first place in all categories, making it the most efficient model on the market to date. It now remains to be determined how this new model will be integrated into the Gemini ecosystem among ordinary users. For the moment, it is only accessible in the Google Ai Studio, at this address.
Advertising, your content continues below
Tech
