Google announced the launch of a new version of its artificial intelligence model, Gemini 2.5 Pro, which is already available to users in Google AI Studio and to Gemini Advanced subscribers. This model promises improved performance thanks to an enhanced base model and better post-training processes. The new version demonstrates high results compared to competitors on platforms such as OpenAI, Anthropic, and others, especially in areas like comprehension, mathematics, programming, and other capabilities.
In addition, Gemini 2.5 Pro stands out for its ability to work with various types of data, including text, audio, images, video, and code. The company also announced that in the future, the model will receive a two-million-token context window, allowing it to process larger volumes of data. According to Google DeepMind CEO Demis Hassabis, the new model has significantly improved its capabilities in multimodal logic, programming, and STEM disciplines.
As part of expanding functionality, Google also plans to introduce a new tool called “Agents,” which will include options such as Deep Research and Canvas. These tools may expand users’ ability to perform complex tasks and create more context-aware agents. Additionally, code references have been found to potential new features such as “AI Transparency Notice” and “Email Opt-in Discovery Card,” indicating Google’s efforts toward transparency and integrating new options for email notification subscriptions.
Google is also working on new features such as Scheduled Prompts, which will allow users to set up periodic execution of queries, as well as a Creative Partner tool that will enhance image editing and video creation capabilities.
