docling-project/docling-serve

Running Docling as an API service

What it solves

It provides a way to run Docling—an AI-driven document conversion toolkit—as a scalable API service. This allows users to integrate document conversion capabilities into other applications via HTTP requests rather than running the toolkit as a local library.

How it works

Docling Serve wraps the Docling conversion engine in a web server (built with FastAPI). It exposes a stable v1 API that accepts document sources (such as URLs) and returns converted content. It can be deployed as a Python package or via container images tailored for different hardware, including CPU-only and various CUDA versions for GPU acceleration.

Who it’s for

Developers and organizations that need to automate document conversion at scale or integrate AI-powered document parsing into a larger software ecosystem.

Highlights

  • API-First Access: Provides a stable v1 API for programmatic document conversion.
  • Flexible Deployment: Offers multiple container images for CPU, CUDA 12.8, and CUDA 13.0, with local build support for AMD ROCm.
  • Interactive UI: Includes a built-in UI playground for testing conversions and visualizing outputs.
  • Easy Integration: Supports standard HTTP POST requests for converting sources like PDFs from URLs.

Related

  • Project
    drmingler/docling-api

    A scalable backend server that uses IBM's Docling to convert various document formats like PDF and DOCX into Markdown via a FastAPI and Celery-based API.

  • Project
    docling-project/docling-mcp

    A document processing service that uses the Model Context Protocol (MCP) to give AI agents tools for converting PDFs into structured data and generating new documents.

  • Project
    docling-project/docling

    A document parsing tool that converts complex PDFs, Office files, and multimodal content into structured Markdown or JSON for use in generative AI and RAG pipelines.

  • Project
    magicrew/doc7

    A document conversion tool that turns PDFs, Office files, and images into AI-ready Markdown using multimodal vision models, supporting local inference for privacy and cost-efficiency.

  • Project
    virgiliojr94/book-to-skill

    A Python tool that converts PDFs, EPUBs, DOCX, Markdown, etc., into an Agent‑Skill bundle (SKILL.md + per‑chapter markdown files). The skill can be loaded by Copilot CLI, Amp, Claude Code, letting the LLM answer questions from the original book on‑demand while using 24‑51× fewer tokens than feeding the whole book as context.