Google launches revamped Gemini Deep Research on Gemini 3 Pro with Interactions API
A new Interactions API lets outside developers embed Google's research agent in their own apps, the first time the tool has been offered outside Google's own products.
- Models & capabilities
- Minor
Google relaunched Gemini Deep Research, its multi-step web research agent, rebuilt on Gemini 3 Pro, and introduced an Interactions API that for the first time lets outside developers embed the agent directly into their own applications rather than accessing it only through Google’s own products. Google positioned the underlying model as its “most factual” yet, aimed at reducing the compounding errors that arise over long, autonomous multi-step research tasks such as due diligence or safety research. The company said it planned to integrate the agent further into Search, Finance, the Gemini app and NotebookLM.
The release landed the same day OpenAI shipped GPT-5.2, and was widely read in the trade press as a deliberately timed counter-move rather than a coincidence. Neither company’s benchmark claims were independently verified at release, and the episode was one of several in December 2025 in which Google and OpenAI shipped competing announcements within hours of each other, underscoring how directly the two companies’ product roadmaps had become reactive to one another following Gemini 3’s benchmark lead the previous month.