In December last year, Google launched the Gemini 2.0 public language model which is more powerful than Gemini 1.5. Gemini 2.0 can also be accessed on the web and applications with the 2.0 Flash model now entering a stable phase. At the same time, the 2.0 Flash Thinking model was also added with the ability to think on its own to solve problems. Last night, 4 Gemini 2.0 models were opened to a number of users.
For now, only Gemini 2.0 Flash has entered a stable phase, Gemini 2.0 Experimental Advanced has been removed and replaced with the previously leaked Gemini 2.0 Pro Experimental as well as Gemini 2.0 Flash Thinking and Gemini 2.0 Flash Thinking with Apps.
Gemini 2.0 Flash promises performance that is many times faster than Gemini 1.5 Pro. It can also solve more complex mathematical problems, in addition to helping with coding tasks. Gemini 2.0 Pro focuses on complex coding tasks with more complex prompts.
For Gemini 2.0 Flash Thinking, it offers a DeepSeek-like usage where Gemini will output shared text about how it thinks and searches for information before generating feedback. It will perform complex thinking and research methods in a short time because it has greater reasoning capabilities.
At the same time, the version with Apps is for this model to be used with applications such as YouTube or Google Maps. This will help users get information faster because this model is an AI agent that helps users use applications.