pip install pymupdf4llm

import pymupdf4llmmd_text = pymupdf4llm.to_markdown("document.pdf")

import pymupdf4llm# 只抽取指定的第8页，第9页md_text = pymupdf4llm.to_markdown("document.pdf", pages=[8, 9])

import pymupdf4llmimport pathlibmd_text = pymupdf4llm.to_markdown("document.pdf")pathlib.Path("output.md").write_bytes(md_text.encode())

import pymupdf4llmllama_reader = pymupdf4llm.LlamaMarkdownReader()llama_docs = llama_reader.load_data("document.pdf")

md_text_images = pymupdf4llm.to_markdown(    doc="document.pdf",    pages=[1, 11],    page_chunks=True,    write_images=True,    image_path="images",    image_format="png",    dpi=300)

md_text_chunks = pymupdf4llm.to_markdown(    doc="document.pdf",    pages=[0, 1, 2],    page_chunks=True)

md_text_words = pymupdf4llm.to_markdown(    doc="document.pdf",    pages=[1, 2],    page_chunks=True,    write_images=True,    image_path="images",    image_format="png",    dpi=300,    extract_words=True)print(md_text_words[0]['words'][:5]) 

import pymupdf4llmmd_text_tables = pymupdf4llm.to_markdown(    doc="document.pdf",    pages=[12]  # 存在表格的页面)print(md_text_tables)