Docling

aid: docling
url: https://raw.githubusercontent.com/api-evangelist/docling/refs/heads/main/apis.yml
apis:
- aid: docling:docling-python-library
  name: Docling Python Library
  tags:
  - Documents
  - Parsing
  - Python
  - SDK
  - PDF
  - OCR
  - LLM
  - RAG
  humanURL: https://docling-project.github.io/docling/
  properties:
  - url: https://docling-project.github.io/docling/
    type: Documentation
  - url: https://docling-project.github.io/docling/getting_started/quickstart/
    type: GettingStarted
  - url: https://github.com/docling-project/docling
    type: SourceCode
  - url: https://pypi.org/project/docling/
    type: SDK
  - url: openapi/docling-cli-openapi.yml
    type: OpenAPI
  - type: NaftikoCapability
    url: capabilities/docling-cli-convert.yaml
  description: 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.
- aid: docling:docling-serve-rest-api
  name: Docling Serve REST API
  tags:
  - Documents
  - Parsing
  - REST
  - PDF
  - OCR
  - Async
  - WebSocket
  humanURL: https://github.com/docling-project/docling-serve
  properties:
  - url: https://github.com/docling-project/docling-serve
    type: Documentation
  - url: https://raw.githubusercontent.com/docling-project/docling-serve/main/docs/usage.md
    name: Docling Serve Usage
    type: Documentation
  - url: https://github.com/docling-project/docling-serve
    type: SourceCode
  - url: openapi/docling-serve-openapi.yml
    type: OpenAPI
  - url: json-schema/docling-document-schema.json
    type: JSONSchema
  - url: json-schema/docling-convert-request-schema.json
    type: JSONSchema
  - url: json-ld/docling-context.jsonld
    type: JSONLD
  - type: NaftikoCapability
    url: capabilities/docling-serve-convert.yaml
  - type: NaftikoCapability
    url: capabilities/docling-serve-tasks.yaml
  description: Docling Serve exposes the Docling pipeline as an HTTP service. Synchronous endpoints `POST /v1/convert/source`
    and `POST /v1/convert/file` accept URL- or upload-sourced documents and return converted JSON, Markdown, HTML, or
    a zipped bundle. Asynchronous variants (`/v1/convert/source/async`, `/v1/convert/file/async`) return a task handle
    that can be polled at `/v1/status/poll/{task_id}`, streamed via WebSocket `/v1/status/ws/{task_id}`, and retrieved
    at `/v1/result/{task_id}`. Container images ship CPU, CUDA 12.8/13.0, and AMD ROCm 6.3 variants; Kubernetes
    deployment is supported via the Docling Operator.
- aid: docling:docling-mcp-server
  name: Docling MCP Server
  tags:
  - MCP
  - Agents
  - Documents
  - Parsing
  humanURL: https://github.com/docling-project/docling-mcp
  properties:
  - url: https://github.com/docling-project/docling-mcp
    type: Documentation
  - url: https://github.com/docling-project/docling-mcp
    type: SourceCode
  description: Model Context Protocol server that exposes Docling document parsing as MCP tools so Claude, Cursor,
    Gemini, and other MCP-aware agents can convert PDFs, Office files, and images into structured `DoclingDocument`
    output without bespoke integration code.
- aid: docling:docling-core
  name: Docling Core Types
  tags:
  - Documents
  - Schema
  - Python
  - SDK
  humanURL: https://github.com/docling-project/docling-core
  properties:
  - url: https://github.com/docling-project/docling-core
    type: Documentation
  - url: https://github.com/docling-project/docling-core
    type: SourceCode
  - url: https://pypi.org/project/docling-core/
    type: SDK
  description: Canonical `DoclingDocument` data model and serialization primitives — text, tables, pictures, layout,
    hierarchy, bounding boxes, provenance — shared by the Docling library, Docling Serve, the Java port, and the
    TypeScript bindings.
- aid: docling:docling-parse
  name: Docling Parse PDF Extractor
  tags:
  - PDF
  - Parsing
  - C++
  humanURL: https://github.com/docling-project/docling-parse
  properties:
  - url: https://github.com/docling-project/docling-parse
    type: Documentation
  - url: https://github.com/docling-project/docling-parse
    type: SourceCode
  description: Native C++ PDF parsing engine used by Docling to extract text with precise coordinates from programmatic
    (non-scanned) PDF files. Distributed as a Python extension.
- aid: docling:docling-ibm-models
  name: Docling IBM Models
  tags:
  - AI
  - Documents
  - Layout
  - TableFormer
  - VLM
  humanURL: https://github.com/docling-project/docling-ibm-models
  properties:
  - url: https://github.com/docling-project/docling-ibm-models
    type: Documentation
  - url: https://github.com/docling-project/docling-ibm-models
    type: SourceCode
  description: Open-weight IBM Research models that power Docling's understanding pipeline — DocLayout (layout
    detection and reading order), TableFormer (table structure), code- and formula-recognition heads, picture
    classifier, and GraniteDocling-258M VLM. Distributed through Hugging Face.
- aid: docling:docling-eval
  name: Docling Eval
  tags:
  - Evaluation
  - Documents
  - Benchmarks
  humanURL: https://github.com/docling-project/docling-eval
  properties:
  - url: https://github.com/docling-project/docling-eval
    type: Documentation
  - url: https://github.com/docling-project/docling-eval
    type: SourceCode
  description: End-to-end evaluation framework for document parsing models and services. Provides standard datasets
    and metrics for layout, tables, OCR, and reading-order quality so teams can benchmark Docling — and competing
    parsers — apples to apples.
- aid: docling:docling-sdg
  name: Docling Synthetic Data Generation
  tags:
  - Synthetic Data
  - Training
  - Documents
  humanURL: https://github.com/docling-project/docling-sdg
  properties:
  - url: https://github.com/docling-project/docling-sdg
    type: Documentation
  - url: https://github.com/docling-project/docling-sdg
    type: SourceCode
  description: Tools for synthesizing labeled document data from real corpora — useful for fine-tuning layout, table,
    and reading-order models, and for stress-testing downstream RAG pipelines.
- aid: docling:docling-graph
  name: Docling Graph
  tags:
  - Knowledge Graph
  - RAG
  - Documents
  humanURL: https://github.com/docling-project/docling-graph
  properties:
  - url: https://github.com/docling-project/docling-graph
    type: Documentation
  - url: https://github.com/docling-project/docling-graph
    type: SourceCode
  description: Transform unstructured documents — once normalized to `DoclingDocument` — into validated, rich, queryable
    knowledge graphs. Intended for GraphRAG and entity-extraction workflows on top of Docling output.
- aid: docling:docling-agent
  name: Docling Agent
  tags:
  - Agents
  - Documents
  - LLM
  humanURL: https://github.com/docling-project/docling-agent
  properties:
  - url: https://github.com/docling-project/docling-agent
    type: Documentation
  - url: https://github.com/docling-project/docling-agent
    type: SourceCode
  description: Reference agent that reads, writes, and edits documents using Docling as the IO layer. Demonstrates how
    Docling output composes with tool-using LLMs to produce structured edits.
- aid: docling:docling-operator
  name: Docling Kubernetes Operator
  tags:
  - Kubernetes
  - Operator
  - Documents
  humanURL: https://github.com/docling-project/docling-operator
  properties:
  - url: https://github.com/docling-project/docling-operator
    type: Documentation
  - url: https://github.com/docling-project/docling-operator
    type: SourceCode
  description: Go-based Kubernetes operator that deploys and manages Docling Serve workloads — model cache PVCs,
    GPU/CPU pools, RQ workers, replica sets with sticky sessions, OAuth — from a single CR.
- aid: docling:docling-java
  name: Docling Java Bindings
  tags:
  - Java
  - SDK
  humanURL: https://github.com/docling-project/docling-java
  properties:
  - url: https://github.com/docling-project/docling-java
    type: Documentation
  - url: https://github.com/docling-project/docling-java
    type: SourceCode
  description: A Java API for Docling that lets JVM applications call into the Docling pipeline. Complementary to
    `docling4j`, which targets Java-native document understanding integrations.
- aid: docling:docling4j
  name: Docling4j
  tags:
  - Java
  - SDK
  humanURL: https://github.com/docling-project/docling4j
  properties:
  - url: https://github.com/docling-project/docling4j
    type: Documentation
  - url: https://github.com/docling-project/docling4j
    type: SourceCode
  description: Brings Docling document understanding into Java projects with idiomatic Java APIs over the Docling
    serialization format.
- aid: docling:docling-ts
  name: Docling TypeScript
  tags:
  - TypeScript
  - JavaScript
  - SDK
  humanURL: https://github.com/docling-project/docling-ts
  properties:
  - url: https://github.com/docling-project/docling-ts
    type: Documentation
  - url: https://github.com/docling-project/docling-ts
    type: SourceCode
  description: TypeScript/JavaScript types and helpers for consuming Docling output (DoclingDocument JSON, DocTags)
    in Node.js and browser applications.
- aid: docling:docling-langchain
  name: Docling LangChain Integration
  tags:
  - LangChain
  - RAG
  - Documents
  humanURL: https://github.com/docling-project/docling-langchain
  properties:
  - url: https://github.com/docling-project/docling-langchain
    type: Documentation
  - url: https://github.com/docling-project/docling-langchain
    type: SourceCode
  description: First-party LangChain document loader and chunker for Docling. Drops Docling output directly into
    LangChain retrieval pipelines.
- aid: docling:docling-jobkit
  name: Docling Jobkit
  tags:
  - Jobs
  - Async
  - Documents
  humanURL: https://github.com/docling-project/docling-jobkit
  properties:
  - url: https://github.com/docling-project/docling-jobkit
    type: Documentation
  - url: https://github.com/docling-project/docling-jobkit
    type: SourceCode
  description: Shared job-runner primitives used by Docling Serve and the Docling Operator to dispatch conversion
    work across RQ workers and Ray.
name: Docling
tags:
- Documents
- Parsing
- PDF
- OCR
- Layout
- Tables
- RAG
- LLM
- Open Source
- IBM Research
- LF AI and Data
- MCP
- Knowledge Graph
- Generative AI
kind: contract
image: https://kinlane-productions2.s3.amazonaws.com/apis-json/apis-json-logo.jpg
access: Open Source
common:
- type: Portal
  url: https://docling-project.github.io/docling/
- type: Documentation
  url: https://docling-project.github.io/docling/
- type: GettingStarted
  url: https://docling-project.github.io/docling/getting_started/quickstart/
- type: SourceCode
  url: https://github.com/docling-project/docling
- type: GitHubOrganization
  url: https://github.com/docling-project
- type: License
  url: https://github.com/docling-project/docling/blob/main/LICENSE
- type: SDK
  url: https://pypi.org/project/docling/
  name: docling on PyPI
- type: SDK
  url: https://pypi.org/project/docling-core/
  name: docling-core on PyPI
- type: SDK
  url: https://pypi.org/project/docling-serve/
  name: docling-serve on PyPI
- type: SDK
  url: https://github.com/docling-project/docling-java
  name: Java bindings
- type: SDK
  url: https://github.com/docling-project/docling4j
  name: Docling4j
- type: SDK
  url: https://github.com/docling-project/docling-ts
  name: TypeScript / JavaScript
- type: CLI
  url: https://docling-project.github.io/docling/reference/cli/
- type: ReleaseNotes
  url: https://github.com/docling-project/docling/releases
- type: ChangeLog
  url: https://github.com/docling-project/docling/blob/main/CHANGELOG.md
- type: Issues
  url: https://github.com/docling-project/docling/issues
- type: Discussions
  url: https://github.com/docling-project/docling/discussions
- type: ContributionGuide
  url: https://github.com/docling-project/docling/blob/main/CONTRIBUTING.md
- type: CodeOfConduct
  url: https://github.com/docling-project/docling/blob/main/CODE_OF_CONDUCT.md
- type: Governance
  url: https://lfaidata.foundation/projects/docling/
  name: LF AI and Data Foundation project page
- type: Foundation
  url: https://lfaidata.foundation/
  name: LF AI and Data Foundation
- type: Models
  url: https://huggingface.co/ds4sd
  name: IBM DS4SD on Hugging Face
- type: Models
  url: https://huggingface.co/ibm-granite/granite-docling-258M
  name: GraniteDocling-258M
- type: Blog
  url: https://research.ibm.com/blog/docling-generative-AI
  name: IBM Research blog — Docling
- type: AcademicPaper
  url: https://arxiv.org/abs/2408.09869
  name: Docling Technical Report
- type: Integration
  url: https://docling-project.github.io/docling/integrations/langchain/
  name: LangChain
- type: Integration
  url: https://docling-project.github.io/docling/integrations/llamaindex/
  name: LlamaIndex
- type: Integration
  url: https://docling-project.github.io/docling/integrations/haystack/
  name: Haystack
- type: Integration
  url: https://docling-project.github.io/docling/integrations/crewai/
  name: Crew AI
- type: Integration
  url: https://docling-project.github.io/docling/integrations/txtai/
  name: txtai
- type: Integration
  url: https://docling-project.github.io/docling/integrations/spacy/
  name: spaCy
- type: Integration
  url: https://docling-project.github.io/docling/integrations/apify/
  name: Apify
- type: Integration
  url: https://docling-project.github.io/docling/integrations/nvidia/
  name: NVIDIA NIM / NeMo Retriever
- type: Integration
  url: https://docling-project.github.io/docling/integrations/instructlab/
  name: InstructLab
- type: Integration
  url: https://docling-project.github.io/docling/integrations/bee/
  name: Bee Agent Framework
- type: Integration
  url: https://docling-project.github.io/docling/integrations/weaviate/
  name: Weaviate
- type: Integration
  url: https://docling-project.github.io/docling/integrations/qdrant/
  name: Qdrant
- type: Integration
  url: https://docling-project.github.io/docling/integrations/milvus/
  name: Milvus
- type: Integration
  url: https://docling-project.github.io/docling/integrations/opensearch/
  name: OpenSearch
- type: ContainerImage
  url: https://quay.io/repository/docling-project/docling-serve
  name: docling-serve container (Quay)
- type: ContainerImage
  url: https://github.com/docling-project/docling-serve/pkgs/container/docling-serve
  name: docling-serve container (GHCR)
- type: KubernetesOperator
  url: https://github.com/docling-project/docling-operator
- type: Features
  data:
  - Parses PDF, DOCX, PPTX, XLSX, HTML, PNG/TIFF/JPEG, WAV/MP3, WebVTT, LaTeX, and plain text
  - Unified DoclingDocument representation with lossless JSON, Markdown, HTML, DocTags, and WebVTT exports
  - Advanced PDF understanding — page layout, reading order, table structure, code, formulas, image classification
  - TableFormer model for accurate table structure recognition
  - GraniteDocling-258M visual language model pipeline for image-first document understanding
  - OCR engines — EasyOCR, Tesseract, RapidOCR, Mac OCR — with per-language configuration
  - Automatic Speech Recognition (ASR) for audio inputs (WAV, MP3) producing WebVTT
  - Local, air-gapped execution — no data leaves the host
  - MCP server (docling-mcp) exposes parsing as agent tools for Claude, Cursor, Gemini and other clients
  - Docling Serve HTTP API with sync and async endpoints, WebSocket task streaming, and zip-bundle output
  - Kubernetes-native deployment via the Docling Operator (model-cache PVCs, RQ workers, GPU pools, OAuth, sticky sessions)
  - Plug-and-play integrations with LangChain, LlamaIndex, Haystack, Crew AI, txtai, Bee, spaCy
  - Application-specific XML schemas (USPTO, JATS, XBRL)
  - Knowledge-graph extraction via docling-graph
  - Synthetic data generation via docling-sdg for fine-tuning
  - End-to-end evaluation framework (docling-eval) with standard datasets and metrics
  - Java, Java-native, TypeScript, and Swift (docling-snap) bindings
  - Open-source MIT license, governed by the LF AI and Data Foundation
  - Originated at IBM Research Zurich (AI for Knowledge team)
  sources:
  - https://docling-project.github.io/docling/
  - https://github.com/docling-project/docling
  - https://github.com/docling-project/docling-serve
  - https://github.com/docling-project/docling-mcp
  - https://lfaidata.foundation/projects/docling/
  - https://arxiv.org/abs/2408.09869
  updated: '2026-05-25'
created: '2026-05-25T00:00:00.000Z'
modified: '2026-05-25'
position: Consuming
description: Docling is an open-source toolkit for parsing diverse document formats — PDF, DOCX, PPTX, XLSX, HTML,
  images, audio, LaTeX, plain text — into a unified, lossless DoclingDocument representation that downstream
  generative AI and RAG systems can consume directly. It pairs IBM Research's DocLayout and TableFormer models with
  the GraniteDocling visual language model and pluggable OCR engines, runs entirely locally for air-gapped use, and
  ships as a Python library and CLI, a FastAPI HTTP service (docling-serve), an MCP server (docling-mcp), and a
  Kubernetes operator. Originally created by IBM Research Zurich; now hosted by the LF AI and Data Foundation under
  the MIT license.
maintainers:
- FN: Kin Lane
  email: [email protected]
  X: apievangelist
  url: https://apievangelist.com
specificationVersion: '0.16'