English | 中文文档
This repo implrements an easy-deployed assistant based on the API of LLMs (OpenAI, Gemma, Anthropic) that can help you to understand pdf files, especially for scholar paper, supporting English, Chinese and multuple languages. We provide a web UI and demo.
- Support long context windows
- Support multiple languages
- Support multiple API (Gemma, Anthropic, etc)
- Improve performance
- More interactive operations
Create an environment and install all dependencies.
conda create -n pdf_assistant python=3.10
conda activate pdf_assistant
pip install -r requirements.txt
Deploy UI by streamlit
streamlit run main.py