Scholar AI is a GenAI-based research assistant that allows you to upload academic PDFs or search arXiv papers and ask questions about them. It uses powerful embeddings and language models to fetch relevant context and provide detailed answers to your research queries.
- β Upload and analyze multiple research paper PDFs
- π Ask questions and get detailed AI-generated responses based on the content
- π§ Uses Google Generative AI (Gemini) for both embeddings and chat-based reasoning
- π Search academic research papers from arXiv by topic
- πΎ Locally stores and retrieves documents using FAISS for fast vector search
- Python
- Streamlit (for the web interface)
- LangChain (for chaining models and prompts)
- Google Generative AI (
gemini-pro,embedding-001) - FAISS (for vector database)
- PyPDF2 (to read PDF content)
- arxiv (to search academic papers)
scholar-ai/ β βββ app.py # Main Streamlit app βββ .env # Contains GOOGLE_API_KEY βββ faiss_index/ # Folder to store vector database βββ requirements.txt # Python dependencies