Google launches Deep Research and Deep Research Max agents to automate complex research

Google has added Deep Research and Deep Research Max to the Gemini API in public preview.

Built on Gemini 3.1 Pro, the agents can automate research using the open web, MCP-connected proprietary data, and multimodal inputs.

The standard version is tuned for faster, lower-cost research, while the Max version targets deeper reasoning and longer-running workflows.

They can return source-backed reports and native charts through a single API call, helping speed up analyst and due-diligence work.