Google expands Gemini features, brings advanced AI tools to more users

Google has announced a set of new updates to its Gemini AI features, aiming to make advanced tools more accessible. The updates, detailed in a blog post by the company, are designed to improve how users interact with artificial intelligence (AI) in everyday tasks, including research, planning and personal productivity.

Flash Thinking model gets upgrades

Gemini’s 2.0 Flash Thinking Experimental model now supports file uploads and offers a longer context window of up to 1 million tokens for advanced users. This allows the model to handle more complex and detailed queries more efficiently.


Deep Research expands globally

The Deep Research tool is now rolling out in over 45 languages. It helps users gather and analyse web data, and now generates more detailed reports with improved reasoning.

Earlier available to select users, it has now been made accessible to free-tier users a few times a month, while Advanced users get extended access.

More personalised responses

Gemini will now offer personalised replies based on a user’s Google Search history—helping with tasks like restaurant suggestions or travel planning. Users can turn off this feature at any time in their settings.

More Google app integrations

Gemini now works more closely with Google Calendar, Tasks, Notes, and soon, Google Photos. Users can now do multi-step actions in one prompt—for example, finding a recipe, making a shopping list, and locating nearby grocery stores.

Soon, it will also create travel itineraries from past trips or extract details from stored documents.

Gems now free for all users

Google is also making Gems, its custom AI assistant feature, free for all users. Users can build their own assistants—like a translator or fitness planner—through a simple setup process.