Google has finally released a new reasoning AI model for its users. The Gemini 2.0 Flash Thinking model is trained in such a way that it breaks down the prompt into multiple steps to improve its reasoning ability and provide more accurate answers. The model also shows its thinking process, allowing users to understand why it answered in a particular way, what were its assumptions and what was its logical chain.
Google Gemini 2.0 Flash Thinking
This launch comes at a time when OpenAI introduced the o3 and o3-mini models of its Reasoning series last week, although Google is a little late in this race, However, the company claims that Gemini 2.0 Flash Thinking is currently the best AI model in the world.
From now on, this model will be available in the model dropdown menu in both desktop and mobile applications. It was launched in December 2024 and is specifically designed to compete with advanced AI systems like OpenAI’s o1 and DeepSeek’s R1.
Gemini 2.0 Flash Thinking experiments with apps
Apart from this, Google is also introducing a new variant “2.0 Flash Thinking Experimental with Apps”. This AI can not only do reasoning but can also interact with services like YouTube, Google Search, and Google Maps. Through these integrations, Google plans to provide a more dynamic and personalized experience to users, making Gemini a powerful AI assistant.
Gemini 2.0 is now available to everyone. ✨
⚡ Start using an updated 2.0 Flash in @Google AI Studio, @GoogleCloud’s #VertexAI and in @GeminiApp.
We’re also introducing:
🔵 2.0 Pro Experimental, which excels at coding.
🔵 2.0 Flash-Lite, our most cost-efficient model yet.
🔵… pic.twitter.com/KbqlteWjSk
— Google DeepMind (@GoogleDeepMind) February 5, 2025
2.0 Pro Experimental is our best model yet for coding and complex prompts, refined with your feedback. 🤝
It has a better understanding of world-knowledge and comes with our largest context window yet of 2 million tokens – meaning it can analyze large amounts of information. pic.twitter.com/DR7S51PJsT
— Google DeepMind (@GoogleDeepMind) February 5, 2025
We’re also excited to reveal 2.0 Flash-Lite. ⚡
It has better quality than 1.5 Flash, at similar cost and speed, and comes with:
🚀 A 1 million token context window
🖼️ Multimodal input
📝 Text output
Now available in @Google AI Studio and @GoogleCloud’s #VertexAI platform in… pic.twitter.com/TVMUdqG8mG
— Google DeepMind (@GoogleDeepMind) February 5, 2025
We’re excited for this next chapter in the Gemini era.
Please keep your feedback coming on our models – and happy building! 🛠️ https://t.co/I2XsTNQpgq pic.twitter.com/BtKNDOWxR7
— Google DeepMind (@GoogleDeepMind) February 5, 2025
Both of these models are now available to all Gemini users on web and mobile platforms. The big advantage of Gemini 2.0 Flash Thinking is that it solves any problem by breaking it down into smaller steps. This ensures that answers are more accurate and reliable, although this method may result in slightly longer response times.
ALSO READ: Asus Zenfone 12 Ultra With 6.78-Inch Display Launched: Top Features, Prices In Taiwan
