Google unveils Deep Research, powered by Gemini 2.5 Pro, for enhanced research efficiency

Google has introduced a new feature for Gemini Advanced subscribers, allowing them to use Deep Research with the Gemini 2.5 Pro Experimental AI model. Gemini Deep Research acts as a personal AI research assistant, designed to aid users throughout the research process. According to Google, in their tests, reports generated by Gemini Deep Research powered by the 2.5 Pro model were preferred over those from other leading deep research providers by a margin of more than 2-to-1. Users have reported improvements in analytical reasoning, information synthesis, and the generation of insightful research reports.

The service is accessible across web platforms, Android, and iOS, enabling users to create detailed, easy-to-read reports on various research topics, significantly reducing the time required for research. Additionally, the Audio Overviews feature allows users to convert their reports into podcast-style audio for convenient listening. To try this feature, users can select Gemini 2.5 Pro (experimental) and tap “Deep Research” in the prompt bar.