Google rolls out new updates to its Gemini app with “thinking” Artifical Intelligence models. The revised Gemini 2.0 AI brings cheaper access for developers, and adds experimental reasoning capability. It is capable of explaining how it answers complex queries to the Gemini app.
As part of Gemini 2.0 AI rollouts, Google has launched the 2.0 Pro and Flash Thinking update. The latter will be accessible via model dropdown options on the desktop and mobile app. The Flash model is touted to be Google’s high-efficiency workhorse AI model.
Gemini 2.0 Pro offers “better factuality” and “stronger performance” for coding and computational tasks, TechCrunch reports citing the leaks of the preview. It is reportedly Google’s “most capable model yet”. Advanced Gemini app users and those running Vertex AI and AI Studio can access Pro and Flash models.
According to The Verge quoting Google, the Flash model’s agentic version also allows direct interaction with apps like YouTube, Search, and Google Maps. Gemini Flash Thinking model launched in Dec 2024. It rivals other reasoning AI models – DeepSeek’s R1 and OpenAI’s o1.
Google recently launched a new affordable model called Gemini 2.0 Flash Lite. The company claims the new Lite matches 1.5 Flash for speed and price, but outperforms it. The Flash Lite is now available in public preview on Google’s AI Studio and Vertex AI.
Leave a Reply