Only after the Google AI Studio and only available to programmers for sophisticated plans subscribers in the Gemini app, The latest version of Google’s AI model with Gemini 2.5 Pro, sophisticated logical capabilities, is available for all customers.
As described by Google, the model part “Family” Gemini 2.5 And for the first time, it is still in experimental mode. Patterns that compose it The ability to “think” before responding to consumer requests, which lead to improved performance and high level accuracyHe confirms the technology.
The company says so “mixing Significantly Advanced Base Model with Optimized Post-Workout Process”He was able to reach New level performance With Gemini 2.5.
Remember Gemini 2.5 Pro has multimodal abilities At the moment, Google is moving forward with the version of the context window with a million tokens.
Google describes Gemini 2.5 Pro Your most sophisticated model for complex tasksHighlighting your performance in the Lmarena ranking, as well as your Logical capabilities and programming areaAs shown in the results obtained in various benchmarks.
According to the company, It was possible to achieve significant improvements in programming with Gemini 2.5 ProWere, and,, and,, and,, and,.. Web applications, including code apps for agents, as well as code transformation and editing. In a previously shared video, Google has already shown how the model can be used Their logical skills Create the game From a single line of code.
