llama-index-readers-file

What it is and what it does

This package provides specialized file parsers for multiple document formats including PDFs, Word documents, spreadsheets, images, notebooks, presentations, and markup files. It wraps several underlying parsing libraries (beautifulsoup4, pypdf, defusedxml, striprtf, and pandas) to handle format-specific extraction. Each loader extracts text, metadata, and structure from its target format and integrates with the document abstraction layer, allowing you to build multi-format ingestion pipelines without writing custom parsing code.

The package is designed as the default integration for file loading, providing readers like DocxReader, PDFReader, EpubReader, HTMLTagReader, ImageReader, IPYNBReader, MarkdownReader, MboxReader, PptxReader, PandasCSVReader, CSVReader, XMLReader, and others. You select the appropriate reader for your file type and pass it to SimpleDirectoryReader, which then loads and structures the content for downstream indexing and retrieval tasks.

Use it for:

• Ingest a directory of mixed PDFs and Word documents for semantic search and indexing.
• Parse CSV files with PandasCSVReader or PagedCSVReader to load tabular data as documents.
• Extract text and structure from HTML files or email archives (MBOX) for knowledge base construction.
• Process Jupyter notebooks or Markdown files to index code and documentation together.
• Load images with ImageReader for multimodal document indexing.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides file loaders that parse documents in multiple formats (PDF, DOCX, images, CSV, HTML, Markdown, and others) into structured data for indexing and retrieval.

Yes. This is actively maintained with low install friction and no known vulnerabilities. It provides broad format coverage through readers for PDFs, DOCX, images, CSV, HTML, Markdown, notebooks, presentations, and more. Install it if you need to ingest documents from multiple file types.

Install

llama-index-readers-file on PyPI

pip install llama-index-readers-file

uv add llama-index-readers-file

poetry add llama-index-readers-file

Installing llama-index-readers-file

Before you install

Low install friction with a pure-Python wheel distribution. Actively maintained with current Python version support.

License in practice

MIT license permits commercial and private use with minimal restrictions.

Quickstart

pip install llama-index-readers-file

from llama_index.readers.file import PDFReader

parser = PDFReader()
file_extractor = {".pdf": parser}
documents = SimpleDirectoryReader("./data", file_extractor=file_extractor).load_data()

Requires llama-index-core as a runtime dependency; Python 3.10 or later.

Verify before relying

• Whether all listed loaders are fully functional or if some require additional system libraries.
• Performance characteristics when processing large files or directories with many documents.
• Whether image-based loaders require additional model downloads or API keys.

Package facts

Evidence: llama_index_readers_file-0.6.0-py3-none-any.whl

Keywords: 10k, 10q, chart, eml, figure, html, hwp, image, invoice, ipynb, jupyter, notebook, pdf, pymupdf, receipt, sec, spreadsheet, tabular, unstructured.io, yaml, yml

Tags

More Text Processing packages