Google has announced the introduction of Gemini 2.0 experimental models, with the Gemini 2.0 Flash Thinking model available for free use. This model enhances reasoning capabilities by breaking down prompts and offering transparency in how decisions are made. Despite claims of 'thinking', Google clarifies that AI doesn't actually reason. Meanwhile, Gemini Advanced subscribers can access Gemini 2.0 Pro, which excels in performing complex tasks with better factuality. These advancements are rolling out across mobile and web platforms, with future availability for Google Workspace users.
Gemini 2.0 Flash Thinking allows users to see the AI's thought process, breaking down prompts into steps to enhance reasoning capabilities and response accuracy.
Google emphasizes that none of its AI models truly 'think' or 'reason', but the new models improve transparency and help users understand AI responses better.
Gemini 2.0 Pro is designed for complex tasks, boasting better factuality, which highlights its improvement in reducing inaccuracies in coding and math tasks.
Both Gemini 2.0 models are rolling out to users, showcasing advances in AI reasoning and performance for a more robust experience in Google's technologies.
Collection
[
|
...
]