I’ve found a good open-source PDF-to-Markdown conversion model—MinerU. It has gained over 28.7k stars on GitHub and can easily convert documents, including Word, PDF, and PPT, to Markdown. It also processes LaTeX, tables, and images with high accuracy. You can deploy it on your PC via Conda, use its GUI software, or even access it through a website.

Click here to get the repositories（https://github.com/opendatalab/MinerU）