$git clone https://github.com/papersgpt/papersgpt-for-zoteroPapersGPT is a professional-grade Zotero extension engineered in Native C++ to transform thousands of isolated PDFs into a single, high-precision knowledge engine. While standard tools struggle with single-file chat or slow, "fuzzy" embeddings, PapersGPT delivers instant, cro
| 1 | # PapersGPT |
| 2 | ## Industrial-Scale Multi-Document Intelligence for Zotero. 100% Local. Zero Hallucinations. |
| 3 | |
| 4 | **PapersGPT** is a professional-grade Zotero extension engineered in Native C++ to transform thousands of isolated PDFs into a single, high-precision knowledge engine. While standard tools struggle with single-file chat or slow, "fuzzy" embeddings, PapersGPT delivers instant, cross-document synthesis and global library search with absolute data sovereignty. |
| 5 | |
| 6 | --- |
| 7 | |
| 8 | ## ⚡ Extreme Performance: Built for "Big Data" Researchers |
| 9 | |
| 10 | Stop waiting for embedding progress bars. Most AI tools use Python/JS scripts that lag or crash as your library grows. **PapersGPT’s C++ core provides:** |
| 11 | |
| 12 | - **Near-Instant Indexing**: Process 1,000+ documents in minutes, not hours. |
| 13 | - **Hardware Efficiency**: Millisecond-level retrieval with minimal RAM usage, leaving your system resources free for other tasks. |
| 14 | - **Zero-Cloud Dependency**: High-speed structural indexing occurs 100% on your machine. No external APIs, no cloud bottlenecks, even works in Airplane Mode. |
| 15 | |
| 16 | --- |
| 17 | |
| 18 | ## 🎯 Superior Multi-Doc Intelligence & Accuracy |
| 19 | |
| 20 | PapersGPT moves beyond the "fuzzy" matching of standard semantic search by embeddings. Our engine understands document structure, ensuring that cross-paper comparisons are logically sound and factually accurate. |
| 21 | |
| 22 | - **Deep Cross-Document Comparison** |
| 23 | *“Compare the sample sizes and p-values of all 50 clinical trials in my 'Cardiology' folder.”* |
| 24 | AI synthesizes data from multiple files into structured tables or reports without losing the nuances of individual studies. |
| 25 | |
| 26 | - **Global Library Synthesis** |
| 27 | Ask questions across your entire Zotero library. Identify trends, evolution of terms, or conflicting evidence across 10,000+ documents simultaneously. |
| 28 | |
| 29 | - **Verifiable Evidence (Click-to-Jump)** |
| 30 | Accuracy is useless without proof. Every synthesis includes interactive citations. Every AI response includes precise citations; clicking on them takes you directly to the specific paper being cited. |
| 31 | |
| 32 | - **Ensure the absolute 100% privacy and security of your local sensitive Zotero data** |
| 33 | |
| 34 | - **Support 100% offline local LLMs** |
| 35 | PapersGPT currently supports three different ways to connect to local large language models (LLMs): |
| 36 | |
| 37 | 1. **Integrated SOTA open-source LLMs** – such as Gemma 4 12B and Qwen3.5 4B. These models can be downloaded and run with a single click, eliminating the need for complex installation procedures; simply selecting a model from the "Local LLM" option triggers an automatic download from Hugging Face and launches the model. |
| 38 | |
| 39 | 2. **Custom OpenAI-compatible endpoint** – connect to a flexible OpenAI-compatible privately deployed LLM endpoint by configuring a customized LLM API within PapersGPT. |
| 40 | |
| 41 | 3. **Ollama integration** – supports connections to local Ollama instances; any LLM deployed and running locally via Ollama can be connected and used simply by configuring it within PapersGPT. |
| 42 | |
| 43 | - **Parsing, Indexing and Searching** documents in Zotero are performed locally, there is no reliance on the clouds, and these operations can be run offline. |
| 44 | |
| 45 | - **Local LLMs and local document processing** ensure **"zero-byte" data leakage** from your device. |
| 46 | |
| 47 | --- |
| 48 | |
| 49 | ## ✨ Core Unique Features |
| 50 | |
| 51 | - **AutoPilot Automation**: The world's first "Autonomous Researcher" for Zotero. Define a research goal and let AI batch-process 1,000+ papers overnight, saving insights directly into Zotero Notes while you sleep. |
| 52 | - **SOTA Model Cowork**: Seamlessly switch between Claude Fable 5, Claude Sonnet 5, GPT-5.6, Gemini 3.6, GLM 5.2, Kimi k3 and DeepSeek, or run Free Local LLMs (via Ollama) for a total air-gapped experience. |
| 53 | - **Zotero MCP Server**: Integrate your local knowledge base as a primary "skill" for agentic platforms like Claude Code, Cursor, or Windsurf. |
| 54 | |
| 55 | --- |
| 56 | |
| 57 | ## 📖 Quickstart |
| 58 | |
| 59 | ### Step 1: Download and Install |
| 60 | |
| 61 | 1. First, download the latest [PapersGPT](https://github.com/papersgpt/papersgpt-for-zotero/releases/download/papersgpt-v0.6.1/papersgpt-v |