Docling

Convert PDFs, DOCX, HTML, images, and other documents to multiple formats like Markdown, JSON, HTML, and more.

DoclingTools enable an Agent to convert documents from multiple input formats (PDF, DOCX, PPTX, XLSX, HTML, images, audio, video, etc.) into output formats like Markdown, JSON, YAML, HTML, DocTags, and more using the Docling library.

Prerequisites

The following example requires the docling library.

uv pip install -U docling

# Required for the OCR example
uv pip install -U easyocr

# Required for audio/video processing
uv pip install -U openai-whisper

ffmpeg is also required for audio/video processing:

• macOS: brew install ffmpeg
• Ubuntu: sudo apt-get install ffmpeg
• Windows: Download from ffmpeg.org

Example

The following agent converts a PDF to Markdown:

from agno.agent import Agent
from agno.tools.docling import DoclingTools

agent = Agent(
    tools=[DoclingTools(all=True)],
    description="You are an agent that converts documents from all Docling parsers and exports to all supported output formats.",
)

agent.print_response(
    "Convert to Markdown: cookbook/07_knowledge/testing_resources/cv_1.pdf",
    markdown=True,
)

OCR Configuration

Configure OCR settings for scanned PDFs or documents with embedded images.

cookbook/91_tools/docling_tools/ocr_example.py

from agno.agent import Agent
from agno.tools.docling import DoclingTools

ocr_tools = DoclingTools(
    pdf_enable_ocr=True,
    pdf_ocr_engine="easyocr",
    pdf_ocr_lang=["pt", "en"],
    pdf_force_full_page_ocr=True,
    pdf_enable_table_structure=True,
    pdf_enable_picture_description=False,
    pdf_document_timeout=120.0,
)

ocr_agent = Agent(
    tools=[ocr_tools],
    description="You are an agent that converts PDFs using advanced OCR.",
)

ocr_agent.print_response(
    "Convert to Markdown: cookbook/07_knowledge/testing_resources/cv_1.pdf",
    markdown=True,
)

Toolkit Params

Parameter	Type	Default	Description
converter	Optional[DocumentConverter]	None	Pre-configured Docling DocumentConverter instance
max_chars	Optional[int]	None	Maximum characters in output
allowed_input_formats	Optional[List[str]]	None	Restrict accepted input formats (e.g. ["pdf", "docx"])
format_options	Optional[Dict[Any, Any]]	None	Custom format options passed to the converter
pdf_pipeline_options	Optional[PdfPipelineOptions]	None	Full PDF pipeline configuration object
pdf_enable_ocr	Optional[bool]	None	Enable OCR processing for PDFs
pdf_ocr_engine	Optional[str]	None	OCR engine: auto, easyocr, tesseract, tesseract_cli, ocrmac, rapidocr
pdf_ocr_lang	Optional[List[str]]	None	OCR language codes (e.g. ["en", "pt"])
pdf_force_full_page_ocr	Optional[bool]	None	Force OCR on every page regardless of text layer
pdf_enable_table_structure	Optional[bool]	None	Enable table structure recognition in PDFs
pdf_enable_picture_description	Optional[bool]	None	Enable picture description extraction
pdf_enable_picture_classification	Optional[bool]	None	Enable picture classification
pdf_document_timeout	Optional[float]	None	Timeout in seconds for PDF processing
pdf_enable_remote_services	Optional[bool]	None	Enable remote services for PDF processing
enable_convert_to_markdown	bool	True	Register the convert_to_markdown function
enable_convert_to_text	bool	True	Register the convert_to_text function
enable_convert_to_html	bool	True	Register the convert_to_html function
enable_convert_to_html_split_page	bool	True	Register the convert_to_html_split_page function
enable_convert_to_json	bool	True	Register the convert_to_json function
enable_convert_to_yaml	bool	True	Register the convert_to_yaml function
enable_convert_to_doctags	bool	True	Register the convert_to_doctags function
enable_convert_to_vtt	bool	True	Register the convert_to_vtt function
enable_convert_string_content	bool	True	Register the convert_string_content function
enable_list_supported_parsers	bool	True	Register the list_supported_parsers function
all	bool	False	Enable all conversion functions when set to True

Toolkit Functions

Function	Description
convert_to_markdown	Converts a document (file path or URL) to Markdown. Accepts source, optional headers for URL requests, raises_on_error, max_num_pages, and max_file_size.
convert_to_text	Converts a document to plain text. Same parameters as convert_to_markdown.
convert_to_html	Converts a document to HTML. Same parameters as convert_to_markdown.
convert_to_html_split_page	Converts a document to HTML with page-level splitting. Same parameters as convert_to_markdown.
convert_to_json	Converts a document to JSON. Same parameters as convert_to_markdown.
convert_to_yaml	Converts a document to YAML. Same parameters as convert_to_markdown.
convert_to_doctags	Converts a document to DocTags format. Same parameters as convert_to_markdown.
convert_to_vtt	Converts a document (including audio/video) to VTT subtitle format. Same parameters as convert_to_markdown.
convert_string_content	Converts raw string content (Markdown or HTML) to another format. Accepts content, source_format (default "markdown"), output_format (default "markdown"), and optional name.
list_supported_parsers	Lists all supported Docling input parsers and any active format restrictions.

Developer Resources

• Tools
• Cookbook Examples
• Docling Documentation