Docling

Docling Python Library

The core Docling Python library and `docling` CLI. Parses PDFs, DOCX, PPTX, XLSX, HTML, images (PNG/TIFF/JPEG), audio (WAV/MP3), WebVTT, LaTeX, and plain text into a unified `DoclingDocument` representation that can be exported to Markdown, HTML, lossless JSON, DocTags, and WebVTT. Implements advanced PDF understanding — page layout, reading order, table structure (TableFormer), code and formula recognition, picture classification — plus OCR (EasyOCR, Tesseract, RapidOCR, Mac OCR) and the GraniteDocling visual language model pipeline. Runs locally for air-gapped and sensitive-data use.

Docling Python Library is one of 16 APIs that Docling publishes on the APIs.io network, described by a machine-readable OpenAPI specification.

This API exposes 1 machine-runnable capability that can be deployed as REST, MCP, or Agent Skill surfaces via Naftiko.

Tagged areas include Documents, Parsing, Python, SDK, and PDF. The published artifact set on APIs.io includes API documentation, a getting-started guide, SDKs, an OpenAPI specification, and 1 Naftiko capability spec.

GitHub OpenAPI

Documentation

📖
Documentation
https://docling-project.github.io/docling/
📖
GettingStarted
https://docling-project.github.io/docling/getting_started/quickstart/

Specifications

⚙
OpenAPI
https://raw.githubusercontent.com/api-evangelist/docling/refs/heads/main/openapi/docling-cli-openapi.yml

SDKs

📦
SDK
https://pypi.org/project/docling/

Other Resources

🔗
SourceCode
https://github.com/docling-project/docling
🔗
NaftikoCapability
https://raw.githubusercontent.com/api-evangelist/docling/refs/heads/main/capabilities/docling-cli-convert.yaml

OpenAPI Specification

docling-cli-openapi.yml Raw ↑ 
openapi: 3.1.0
info:
  title: Docling CLI as REST
  description: |
    REST-style mapping of the `docling` command-line tool, modeling each conversion
    invocation as a `POST /v1/convert` call. Useful for documenting and orchestrating
    local CLI runs from agent frameworks and capability-based runtimes.
  version: '1.0'
  license:
    name: MIT
    url: https://github.com/docling-project/docling/blob/main/LICENSE
  contact:
    name: Docling Project
    url: https://github.com/docling-project/docling
servers:
- url: file:///usr/local/bin
  description: Local docling CLI binary.
tags:
- name: Convert
  description: Document conversion operations.
paths:
  /v1/convert:
    post:
      tags:
      - Convert
      summary: Convert A Document Via The Docling CLI
      description: |
        Run the `docling` command against one source (file path or URL) and write the
        configured outputs to disk. Mirrors the `docling [source] [flags]` invocation.
      operationId: cliConvert
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CliConvertRequest'
      responses:
        '200':
          description: Conversion succeeded.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CliConvertResponse'
        '400':
          description: Invalid arguments.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CliError'
        '500':
          description: Conversion failed.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CliError'
components:
  schemas:
    CliConvertRequest:
      type: object
      required:
      - source
      properties:
        source:
          type: string
          description: File path, directory, or HTTP(S) URL to convert.
        from_formats:
          type: array
          items:
            type: string
            enum: [pdf, docx, pptx, xlsx, html, md, asciidoc, image, audio, csv]
          description: Input formats to accept when scanning directories.
        to_formats:
          type: array
          items:
            type: string
            enum: [md, html, json, text, doctags]
          description: Output formats to write.
        output:
          type: string
          description: Output directory.
        ocr:
          type: boolean
        ocr_engine:
          type: string
          enum: [easyocr, tesseract, tesseract_cli, rapidocr, ocrmac]
        ocr_lang:
          type: array
          items:
            type: string
        force_ocr:
          type: boolean
        table_mode:
          type: string
          enum: [fast, accurate]
        pipeline:
          type: string
          enum: [standard, vlm]
        vlm_model:
          type: string
          enum: [granite_docling, granite_docling_mlx, smoldocling]
        image_export_mode:
          type: string
          enum: [embedded, placeholder, referenced]
        enrich_code:
          type: boolean
        enrich_formula:
          type: boolean
        enrich_picture_classes:
          type: boolean
        enrich_picture_description:
          type: boolean
        device:
          type: string
          enum: [auto, cpu, cuda, mps]
        num_threads:
          type: integer
        abort_on_error:
          type: boolean
        verbose:
          type: boolean
    CliConvertResponse:
      type: object
      properties:
        outputs:
          type: array
          items:
            type: object
            properties:
              path:
                type: string
              format:
                type: string
        status:
          type: string
          enum: [success, partial_success, failure]
        processing_time:
          type: number
    CliError:
      type: object
      properties:
        message:
          type: string
        code:
          type: integer