Google has just released the new Gemini 3 model which is intended to replace the Gemini 2.5, which was introduced last May.
Its release Gemini 3 will initially be released with its integration into the “AI Mode" of the Google search engine, as well as through the Gemini application which is available to everyone.
The CEO of Google, Sundar Pichai, reports that the new model is much better at understanding the context and intent behind the user's request, without needing more guidance to produce a response.
Furthermore, the Google says that Gemini 3 is the world's best model for multimodal understanding, providing improved visualizations and enhanced interactivity.
Initially with its release, there is the Gemini 3 Pro and then, the will come Gemini 3 Deep Think, which is still being tested for security issues. Once its trial period is over, it will be available to subscribers of the service Google AI Ultra.
According to AI benchmarks, the new model Gemini 3 Pro is far superior compared to Gemini 2.5 ProAccording to Google, it provides answers to users' questions that are smart, direct, and exchange clichés and flattery with substantive and real insight.
As far as Gemini 3 Deep Think, is superior compared to Gemini 3 Pro according to AI benchmarks, but as a natural and consequential matter, it takes longer with its responses.
Since the new model has been built and developed from the ground up with the ability to synthesize information on any subject/topic through multiple formats, it is the ideal model for users who want to develop their knowledge on a subject.
On the issue of security Gemini 3, Google says it is the most secure model and has undergone the most comprehensive set of security assessments of any other Google AI model. This results in reduced flattery, increased resistance to prompt injections and improved protection against abuse due to cyberattacks.





