Y

+1，Up to now, the processing effect of this project on PDF is still not good, including pictures, tables (including merged cells), titles, etc., are not good

> [@SigireddyBalasai](https://github.com/SigireddyBalasai) Right,I get it now,I had expected to receive the main content,not just a description,thx This may require an OCR approach

> md = MarkItDown(llm_client=client, llm_model="gpt-4o") ,图片转md需要提供llm的接口才能识别 This method can summarize the content of the picture, but can not identify the text in the picture, how to extract the text in...

> ### Description of the bug | 错误描述 > > ### How to reproduce the bug | 如何复现 > 详情见论文《基于平衡计分卡的行政单位预算绩效评价优化研究——以H 市税务局为例》.pdf > > ### Operating system | 操作系统 > MacOS...

> [@Fanzaijun](https://github.com/Fanzaijun) The accepted reporting language of issues in this tracker is English only. Can you please explain what issue you see. Thanks. Apart from Hugging face, can the parameters...