Google DeepMind has launched two new autonomous research agents that can dig through massive amounts of data, connect to your private files, and automatically generate professional reports with charts and visualizations. Called Deep Research and Deep Research Max, these tools represent a significant leap forward in how businesses can gather expert-level analysis at scale. Built on Gemini 3.1 Pro, the company's most advanced model, these agents are designed to handle everything from financial due diligence to market research and life sciences analysis.

What Makes Deep Research Max Different From Earlier Versions?

When Google first released the Gemini Deep Research agent to developers in December, it was primarily a sophisticated summarization engine. The new versions transform that foundation into something far more powerful. Deep Research Max, in particular, uses extended test-time compute to iteratively reason, search, and refine reports, delivering what Google describes as "unprecedented analytical quality". Compared to the December release, Deep Research Max consults significantly more sources and identifies critical nuances that earlier versions frequently overlooked.

The key difference between the two new agents comes down to speed versus comprehensiveness. Deep Research is optimized for fast, interactive experiences where users need quick answers, while Deep Research Max is designed for asynchronous, background workflows like generating exhaustive due diligence reports overnight for analyst teams to review in the morning.

"Deep Research has transformed from a sophisticated summarization engine into a foundation for enterprise workflows across finance, life sciences, market research, and more," stated Lukas Haas, Product Manager at Google DeepMind.

Lukas Haas, Product Manager, Google DeepMind

How to Integrate Deep Research Into Your Workflow

• Connect Custom Data Sources: Use the Model Context Protocol (MCP) to securely link Deep Research to your proprietary data streams, financial databases, or specialized professional data providers without exposing sensitive information to the open web.
• Generate Visual Reports Automatically: The agent now natively creates high-quality charts and infographics in HTML or other formats, turning complex datasets into presentation-ready visualizations without manual design work.
• Control the Research Plan: Review and refine the research plan generated by the agent before execution begins, giving you granular control over the investigation's scope and ensuring the agent focuses on what matters most to your team.
• Combine Multiple Data Sources: Run Deep Research simultaneously with Google Search, remote MCP servers, URL context analysis, code execution, and file search capabilities, or disable web access entirely to search only your custom data.
• Track Real-Time Progress: Monitor the agent's intermediate reasoning steps with live thought summaries and receive text and image outputs as they are generated, which is particularly useful for interactive user interfaces.

Which Industries Are Already Using This Technology?

Google DeepMind is actively collaborating with startups and enterprises in specialized and regulated fields where accuracy is critical. The company is working closely with financial data providers like FactSet, S&P Global, and PitchBook on their MCP server designs. These partnerships allow shared customers to integrate financial data offerings directly into workflows powered by Deep Research, enabling teams to gather context using exhaustive data universes at significantly faster speeds.

The life sciences sector is another major focus area. Because these fields have little margin for error, Google is ensuring Deep Research Max delivers highly comprehensive reports with rigorous factuality and expert-grade analysis. The agent has been trained to consult diverse arrays of sources and carefully weigh conflicting evidence, drawing from authoritative sources like SEC filings and open-access peer-reviewed journals.

When Can You Start Using These Tools?

Deep Research and Deep Research Max are available starting today in public preview via paid tiers in the Gemini API, according to Google's announcement. Developers can access these tools through the Interactions API and begin building research workflows immediately. The tools will also soon be available to startups and enterprises through Google Cloud.

These new agents tap into the same autonomous research infrastructure that powers research capabilities within some of Google's most popular products, including the Gemini App, NotebookLM, Google Search, and Google Finance. This means the underlying technology has already been tested and refined across millions of real-world use cases.

For enterprises drowning in data and analysts spending weeks on research projects, Deep Research Max offers a concrete way to accelerate workflows without sacrificing accuracy or control. The ability to connect to private data sources while maintaining security, combined with automatic visualization generation, addresses a real pain point in how professional teams currently conduct research and analysis.