pdfusion

A privacy-first PDF processing engine that deconstructs documents into their core elements—text, images, and tables—and reconstructs them into pristine, structured Markdown. Self-hosted React + FastAPI stack with local AI vision models via LiteLLM/Ollama. Your data never leaves your machine. Maintained by seehiong on GitHub, where it has earned 68 stars from the community.

It's actively developed around backend, document-processing, fastapi, and is a solid reference for anyone building with these tools.