# Docling — Document Parsing for AI Pipelines

> Parse PDF, DOCX, PPTX, HTML, and images into clean Markdown or JSON for LLM ingestion. Handles tables, figures, equations, and complex layouts. By IBM Research. 18,000+ stars.

## Install

Save as a script file and run:

## Quick Use

```bash
pip install docling
```

```python
from docling.document_converter import DocumentConverter

converter = DocumentConverter()
result = converter.convert("report.pdf")
print(result.document.export_to_markdown())
```

CLI usage:
```bash
docling convert report.pdf --output-format markdown
```

---

## Intro

Docling is a document parsing library by IBM Research that converts PDF, DOCX, PPTX, HTML, and images into clean Markdown or JSON optimized for LLM ingestion with 18,000+ GitHub stars. It handles complex layouts including tables, figures, mathematical equations, code blocks, and multi-column text — elements that naive PDF parsers destroy. Best for teams building RAG pipelines that need high-fidelity document parsing. Works with: any LLM pipeline, Haystack, LangChain, LlamaIndex. Setup time: under 2 minutes.

---

## Supported Formats

| Format | Tables | Figures | Equations | Code |
|--------|--------|---------|-----------|------|
| PDF | Yes | Yes | Yes | Yes |
| DOCX | Yes | Yes | Yes | Yes |
| PPTX | Yes | Yes | N/A | Yes |
| HTML | Yes | Yes | Yes | Yes |
| Images | OCR | Yes | OCR | OCR |
| Markdown | Pass-through | Yes | Yes | Yes |

### Table Extraction
Docling accurately extracts tables preserving structure:

```python
result = converter.convert("financial_report.pdf")
for table in result.document.tables:
    print(table.export_to_dataframe())  # pandas DataFrame
```

### Figure Handling
Figures are extracted with captions and alt-text:

```python
for figure in result.document.pictures:
    print(f"Caption: {figure.caption}")
    figure.image.save(f"figure_{figure.id}.png")
```

### OCR for Scanned Documents
Built-in OCR for scanned PDFs and images:

```python
converter = DocumentConverter(ocr=True)
result = converter.convert("scanned_document.pdf")
```

### Batch Processing

```python
from docling.document_converter import DocumentConverter
from pathlib import Path

converter = DocumentConverter()
docs = Path("./documents").glob("*.pdf")
for doc in docs:
    result = converter.convert(str(doc))
    markdown = result.document.export_to_markdown()
    Path(f"./output/{doc.stem}.md").write_text(markdown)
```

### Integration with RAG

```python
# With Haystack
from docling.integrations.haystack import DoclingConverter
converter = DoclingConverter()
documents = converter.run(sources=["report.pdf"])

# With LangChain
from docling.integrations.langchain import DoclingLoader
loader = DoclingLoader(file_path="report.pdf")
docs = loader.load()
```

### Key Stats
- 18,000+ GitHub stars
- By IBM Research
- 6 input formats supported
- Table, figure, equation extraction
- OCR for scanned documents
- Haystack, LangChain integrations

### FAQ

**Q: What is Docling?**
A: Docling is a document parsing library by IBM Research that converts PDF, DOCX, PPTX, and other formats into clean Markdown or JSON for LLM consumption, handling tables, figures, and complex layouts.

**Q: Is Docling free?**
A: Yes, open-source under MIT license by IBM Research.

**Q: How accurate is Docling compared to PyPDF?**
A: Significantly more accurate for complex documents. Docling uses deep learning models to understand document layout, while PyPDF does simple text extraction.

---

## Source & Thanks

> Created by [DS4SD / IBM Research](https://github.com/DS4SD). Licensed under MIT.
>
> [docling](https://github.com/DS4SD/docling) — ⭐ 18,000+

Thanks to IBM Research for solving document parsing for the AI era.

---

<!-- ZH -->

## 快速使用

```bash
pip install docling
```

```python
from docling.document_converter import DocumentConverter
result = DocumentConverter().convert("report.pdf")
print(result.document.export_to_markdown())
```

---

## 简介

Docling 是 IBM Research 开发的文档解析库，GitHub 18,000+ stars。将 PDF、DOCX、PPTX 等格式转换为干净的 Markdown 或 JSON，精确处理表格、图表、数学公式和复杂版面。适合构建 RAG 管道需要高保真文档解析的团队。

---

## 来源与感谢

> Created by [DS4SD / IBM Research](https://github.com/DS4SD). Licensed under MIT.
>
> [docling](https://github.com/DS4SD/docling) — ⭐ 18,000+

---
Source: https://tokrepo.com/en/workflows/a8327829-385d-47cf-9b22-fa9a5d2aafde
Author: Script Depot