pip install pymupdf4llm

import pymupdf4llm# Extract PDF content as Markdownmd_text = pymupdf4llm.to_markdown("input.pdf")print(md_text[:500])  # Print first 500 characters

import pymupdf4llm# Extract only pages 10 and 11md_text = pymupdf4llm.to_markdown("bengio03a.pdf", pages=[10, 11])print(md_text[:500])  # Print first 500 characters

import pymupdf4llmllama_reader = pymupdf4llm.LlamaMarkdownReader()llama_docs = llama_reader.load_data("bengio03a.pdf")print(f"Number of LlamaIndex documents: {len(llama_docs)}")print(f"Content of first document: {llama_docs[0].text[:500]}")

import pymupdf4llmimport json
md_text_images = pymupdf4llm.to_markdown(doc="bengio03a.pdf",                                         pages=[1, 11],                                         page_chunks=True,                                         write_images=True,                                         image_path="images",                                         image_format="jpg",                                         dpi=200)print(md_text_images[0]['images'])  # Print image information from the first chunk
md_text_tables = pymupdf4llm.to_markdown(doc="bengio03a.pdf",                                         pages=[12],  # Specify pages containing tables                                         )print(md_text_tables)