Developed within NFDIxCS

Extract insights from PDFs
with LLMs in Zotero

ZotExtract is a Zotero 7 plugin that lets you query large language models about your research PDFs. Run local LLMs for full privacy or connect to online providers.

Download Latest Release View on GitHub

Features

Everything you need to leverage LLMs for your research workflow — right inside Zotero.

📄

PDF Context Extraction

Automatically extract full text from PDFs or send the raw PDF file to a file-capable AI endpoint for deep analysis.

📚

Batch Processing

Select multiple items and process them all at once. A progress bar tracks each item as it's analysed by the LLM.

💬

Prompt Library

Use built-in prompts for summaries, methods, and limitations — or create and save your own custom questions.

🔒

Privacy-First

Run local LLMs (Ollama, LM Studio, etc.) for complete data privacy. API keys stay in your local Zotero profile.

📝

Auto-Save as Notes

LLM responses are saved as child notes on each item, keeping your analysis organised and searchable in Zotero.

⚙️

Flexible Configuration

Connect any OpenAI-compatible API. Configure model, base URL, system prompt, and custom PDF endpoints.

Get Started in Minutes

Install the plugin, configure your LLM provider, and start extracting insights.

1

Download

Grab the latest .xpi file from the Releases page on GitHub.

2

Install

In Zotero 7, go to Tools → Add-ons, click the gear icon, and choose Install Add-on From File.

3

Configure

Open Settings → ZotExtract and enter your API base URL, model, and key. Or point to a local Ollama server.

4

Extract

Right-click any item → ZotExtract. Pick a prompt and get AI-powered insights saved as notes.

Works With Your Favourite LLM

Connect to any OpenAI-compatible API — cloud or local.

OpenAI
Ollama
LM Studio
Any OpenAI-compatible API

Funding & Acknowledgements

ZotExtract is developed at the Chair of Information Science, University of Regensburg, as part of NFDIxCS — the National Research Data Infrastructure for and with Computer Science (NFDI×CS). NFDIxCS promotes the implementation of FAIR Data Principles for computer-science research data and software artefacts.

NFDIxCS is funded by the German Research Foundation (DFG) under the National Research Data Infrastructure (NFDI) programme.

nfdixcs.org Mastodon Bluesky LinkedIn
Developed at University of Regensburg — Chair of Information Science Contact: Samy Ateia
NFDIxCS operated by Gesellschaft für Informatik e.V. Consortium led by University of Duisburg-Essen