shibing624/ChatPDF
A local RAG system that enables conversational Q&A over PDF/doc/txt files using open-source LLMs and embedding models.
Not currently ranked — collecting fresh signals.
star history
This project implements a lightweight GraphRAG system for document question answering based on local LLMs. It supports multiple LLM providers including OpenAI, DeepSeek, and Ollama, along with various embedding models like text2vec and sentence-transformers. The system optimizes RAG accuracy through Chinese-aware text chunking, BM25 keyword matching combined with vector similarity search, and a reranker module to refine retrieved results.
Frequently asked
- What is shibing624/ChatPDF?
- A local RAG system that enables conversational Q&A over PDF/doc/txt files using open-source LLMs and embedding models.
- Is ChatPDF open source?
- Yes — shibing624/ChatPDF is open source, released under the Apache-2.0 license.
- What language is ChatPDF written in?
- shibing624/ChatPDF is primarily written in Python.
- How popular is ChatPDF?
- shibing624/ChatPDF has 854 stars on GitHub.
- Where can I find ChatPDF?
- shibing624/ChatPDF is on GitHub at https://github.com/shibing624/ChatPDF.