ByLast year, Google introduced a new feature in Gemini called Deep Research. It is an AI research assistant that digs deep into complex topics and generates reports with reliable sources. In the process, it can browse hundreds of websites, analyze information, and even turn reports into podcast-style conversations.
Last month, Google announced that Gemini Deep Research is now powered by the Gemini 2.0 Flash Thinking Experimental model, which improves its ability to plan, search, reason, analyze, and generate reports. After this upgrade, users started getting better quality, multi-page reports.
Today, Google announced that Deep Research is powered by Gemini 2.5 Pro Experimental, its most advanced model. Gemini Advanced users can now use this feature on the web, Android, and iOS. With this powerful new model, Deep Research is now able to deliver even better planning, discovery, analysis, reasoning, and reporting.
Want to grow your business organically? Contact us now
According to Google’s testing, human raters preferred Gemini 2.5 Pro-powered Deep Research reports more than 2 times more often than OpenAI’s Deep Research. Users also felt that Deep Research’s analytical thinking and ability to combine information was significantly improved after this upgrade, making reports even more informative and in-depth.
If you are a Gemini Advanced user, you can select the “Gemini 2.5 Pro (experimental)” model by going to the model selection dropdown and selecting the “Deep Research” option in the prompt bar. This upgrade is another step in Google’s continued improvement of Gemini Advanced’s capabilities, giving users even more powerful tools for complex research tasks.
You can also try Deep Research with the Gemini 2.0 Flash model for free in 150 countries on desktop and mobile.
