Quick Facts
- Deep Research Max scored 93.3% on DeepSearchQA benchmark, up from 66.1% in December 2025
- New Model Context Protocol integration allows secure access to private databases without data leaving source environment
- Enterprise partnerships with FactSet, S&P Global, and PitchBook enable financial data integration
Google unveiled Deep Research and Deep Research Max on Monday, marking the company’s most significant upgrade to autonomous research capabilities since the product’s debut. The new AI agents can access both open web data and proprietary enterprise information through a single API call.
The agents are powered by Google’s Gemini 3.1 Pro model, released in February 2026. Deep Research Max achieved 93.3% on DeepSearchQA benchmarks, compared to 66.1% for the December 2025 release. The agent also scored 54.6% on Humanity’s Last Exam, up from 46.4%.
The breakthrough feature is Model Context Protocol (MCP) support, which allows Deep Research to query private databases and internal document repositories securely. MCP prevents sensitive information from leaving its source environment while enabling comprehensive data analysis.
Google is partnering with FactSet, S&P Global, and PitchBook to integrate their financial data offerings into Deep Research workflows. These partnerships target finance professionals who need exhaustive data analysis at high speed.
The company offers two tiers: Deep Research for lower-latency interactive use and Deep Research Max for comprehensive background analysis. Deep Research Max can use up to 160 search queries and process 900,000 input tokens for complex research tasks within a 60-minute maximum timeframe.
For the first time in the Gemini API, the agents can generate native charts and infographics directly inside reports. This visual capability addresses enterprise needs for data presentation in regulated industries like finance and healthcare.
“Deep Research was our first hosted agent in the API and has gained a ton of traction over the last 3 months,” wrote Logan Kilpatrick, who leads developer relations for Google’s AI efforts.
The launch puts Google in direct competition with OpenAI’s research agents and Exa’s Deep Max API. The timing reflects intensifying competition in autonomous research agent capabilities across the AI industry.
Both agents are available today in public preview through paid tiers in the Gemini API. Google Cloud availability for startups and enterprises will follow soon. The company is targeting specialized industries where research accuracy is critical, including life sciences and financial services.
Read more: Google’s new Deep Research and Deep Research Max agents can search the web and your private data