Gemini 2.0 Flash Thinking and 2.0 Pro Experimental rolling out to Gemini app

Following the Internet, the Gemini application on Android and iOS begins to access the experimental experimental and 2.0 Pro Experimental (if you are an extended subscriber).

Gemini 2.0 Flash Spearing Experimental Google reasoning model that shows its thoughts. It was represented in December and updated in January. The speed and performance of 2.0 Flash, Gemini breaks the hint into steps that you can see.

On the desktop you will see: show thinking, determine the sphere of question, recognize various prospects, key brainstorming concepts and related questions, structure the answer, clarify and develop in each section, consider the possibility of adding nuances and warnings, review and editing.

In a mobile application, this is a simpler section of “thoughts”, followed by a “answer”.

The Gemini application shows this in real time when users can see the stream text (although it acts faster than you can read). This allows you to understand “why he answered in a certain way, what are his assumptions, and trace the line of reasoning of the model.” You can also press Chevron to hide it. As a preliminary viewing, “there is not enough access to information in real time and some functions of the twins.”

2.0 Flash Speading Experimental will be available for free users, and we see it on several Android devices today.

Google also announced Twins 2.0 Flash -thoughts experimentally with applications For “Reasoning in Google applications.” It “works best” with YouTube, cards and search, ”while some functions of the twins are inaccessible. For comparison, the main model is “best for multi -stage reasoning.”

On mobile devices, some people see “with applications” instead of 2.0 flash -thoughts experimentally.

Meanwhile, Gemini 2.0 Pro Experimental (Gemini-2 ,0-Pro-EXP-02-05) has “the best understanding and reasoning of world knowledge” than any of the previous Google models. They say that this is ideal for complex and encoding tips.

Gemini Advanced has a 1 million token context window, but the developer API offers 2 million. At this stage, he “does not have access to information in real time and some functions of the twins.” Since the deployment continues, paid subscribers seem to get new models to get ahead of free accounts.

After releasing the Gemini App Model Picker model:

• 2.0 flash: For everyday tasks
• 2.0 experimental thinking.: Best for multi -stage reasoning
• 2.0 Flash -thoughts experimentally with applications: Reasoning on YouTube, Mars & Search
• 2.0 Pro Experimental: Best for difficult tasks
• 1.5 Pro with deep research: Get detailed answers
• 1.5 Pro: Previous model
• 1.5 flash: Previous model

Old against the new