Gemini Deep Research
Autonomous web research baked into Gemini, not sold on its own
Overview
Deep Research is Gemini's agentic mode that plans a research approach, browses dozens of sources, and returns a long cited report exportable straight to Google Docs. Since May 2026 its limits are compute-based and scale with the free, AI Plus, Pro and Ultra plans rather than a separate price.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 30 Sep 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Thinking model
- 32K context
- Thinking model
- 128K context
- 400 GB storage
- Reports on the Pro model
- 1M token context
- 5 TB storage
- Reports on the Pro model, with visuals
- Deep Think reasoning
- From 20 TB storage
Use cases
What you can produce with Gemini Deep Research
- Autonomous multi-source web research reports
- One-click export to Google Docs
- Reads your own Gmail, Drive and Chat as sources, if you allow it
- Upload your own files as research sources
- Up to 1M-token context window on AI Pro and Ultra
- Visual reports with charts, diagrams and simulations, plus Deep Think (Ultra)
- A Deep Research agent in the Gemini API (preview)
Deep Research works because it's free to try and fast at first-pass research: point it at a broad question and it comes back in minutes with a structured, cited report and an Export to Docs button, pulling from dozens of web sources plus your own Drive and Gmail if you allow it. Access scales with subscription rather than a standalone price: since May 2026 limits are compute-based — standard on the free tier, about twice that on AI Plus ($4.99/mo in the US), four times on AI Pro ($19.99/mo) and five or twenty times Pro on the two Ultra tiers ($99.99 and $199.99) — and only Pro and Ultra reports run on the Pro model. The catch is citation quality. Threads on Google's own developer forum document it inventing patent numbers and details, and the ReportBench academic benchmark puts its citation alignment behind OpenAI's deep research on the same tasks. Treat every claim as a lead to verify, not a fact to cite downstream.
What we found
Deep Research is Gemini's agentic research mode: give it a broad question and it plans an approach, reads dozens of web sources — plus your Gmail, Drive and Chat if you allow it, and files you upload — and returns a long, cited report in minutes, with an Export to Docs button. On AI Ultra, reports can include charts, diagrams and interactive simulations.
It is not sold on its own in the Gemini app; access scales with Google's AI plans. In May 2026 Google replaced fixed daily limits with compute-based ones: the free tier gets standard limits, AI Plus about twice that, AI Pro four times, and AI Ultra five or twenty times Pro's. Pro and Ultra subscribers can run reports on the Pro model; everyone can use the Thinking model, and free users may find it unavailable at busy times. US prices are $4.99 a month for Plus, $19.99 for Pro, and $99.99 or $199.99 for the two Ultra tiers. Developers can also try a Deep Research agent in the Gemini API, currently in preview.
The catch is citation quality. An independent academic benchmark, ReportBench, found Gemini's cited statements matched their sources about 73% of the time, against about 79% for OpenAI's equivalent, and users on Google's own developer forum have documented it inventing patent numbers and details. Treat every claim as a lead to verify, not a fact to cite downstream.
The strengths are speed and reach: for first-pass research across the open web and your own documents, few tools are as quick to start with or as easy to share, since reports go straight into Google Docs.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including Gemini Deep Research's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Gemini Deep Research unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Gemini Deep Research directly →
Spotted something out of date? Suggest an update →
Related reading
Alternatives to Gemini Deep Research
More in Research & Data
