AI research across your own documents and the open web
Bring what you already know. Citaria fills the gaps from the web and cites both.

In short
Citaria researches your own files and the open web in one research plan, starting from your project files and filling gaps from the web. Citaria reads PDFs with their tables rebuilt, Word documents, Markdown, text, VTT and SRT transcripts, and audio and video recordings, and cites every claim to its passage, whatever the source.
- Files
- PDF, Word (.docx), Markdown, text, VTT, SRT, audio, video
- Size
- Up to 2 GB per file
- Google Drive
- Import Google Docs, Word and text files, and recordings
- Websites
- Trusted sites favored; blocked sites never searched or read
- No project
- Attach up to 20 files to a single question
One plan across every source
- 01
Add what you have
Upload files to a project, import them from Google Drive, or attach them to a single question.
- 02
Research it with the web
Citaria plans the research from your own files out to the open web, closing gaps as they surface.
- 03
See where each claim came from
Every claim opens the passage behind it, so you can tell your files from the web at a glance.
Your context, plus the world’s
Customer research
Combine your interview recordings and notes with public market data in one report.
Strategy reviews
Check your internal plans against what the web says today.
Follow-up research
Build on a report you wrote earlier in Citaria, cited like any other source.
Which AI research tools can research your own files and the web?
Citaria is our product. Facts about other tools come from their official documentation; the review date and sources are at the foot of this page.
| Tool | What it can research |
|---|---|
| Citaria | Your project files, audio and video recordings, and the open web |
| ChatGPT Deep Research | The web, uploaded files, and connected apps such as Google Drive or SharePoint |
| Perplexity | The web, plus files you upload |
| Gemini Deep Research | Google Search, uploaded files, Gmail, Drive, and Google Chat |
| NotebookLM | A notebook of sources you add, with Fast or Deep Research to find more |
| Elicit | Academic papers and clinical trials; the Research Agent also searches the web and your uploads |
Can an AI research my own documents and the web at the same time?
Yes. Citaria researches your project files and the open web in one research plan and cites each claim to its passage, so you can see which findings come from your documents and which come from the web.
Which file types can Citaria research?
PDF, Word (.docx), Markdown, and text documents; VTT and SRT transcripts; and audio and video recordings, up to 2 GB per file. From Google Drive, you can import Google Docs, Word and text files, and recordings.
Can I limit which websites Citaria uses?
Yes. Each project can list trusted websites that Citaria favors in its searches, and blocked websites it never searches or reads.
Is my research material used to train AI models?
No model is trained on your research material. Workspaces are kept separate, and data is encrypted in transit and at rest. The Trust page has the details.