whatisgithub

What is docslight?

compdf/docslight — explained in plain English

Analysis updated 2026-05-18

50VueAudience · developerComplexity · 3/5LicenseSetup · moderate

In one sentence

An open source document parser that turns PDFs, scans, images, and Office files into structured Markdown or JSON using AI parsing and multilingual OCR, runnable locally or via a cloud API.

Mindmap

mindmap
  root((DocSlight))
    What it does
      Parses PDFs and Office files
      OCR for 80 plus languages
      Structured field extraction
    Modes
      Local free mode
      Cloud API mode
      Web UI via Docker
    Tech stack
      Python
      Vue
      Docker
    Use cases
      RAG pipelines
      Invoice processing
      Contract analysis
    Audience
      Developers
      Enterprise IT teams

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Convert a folder of invoices or contracts into structured JSON for downstream processing.

USE CASE 2

Feed parsed document text into a retrieval-augmented generation pipeline or knowledge base.

USE CASE 3

Give an AI agent a tool to read and extract data from PDFs, scans, or Office files.

USE CASE 4

Run a local, no-account document parser instead of sending files to a third-party cloud service.

What is it built with?

PythonVueDockerOCR

How does it compare?

compdf/docslightageerle/ruoyi-dramahuangdihd/call_me_as_agent
Stars505050
LanguageVueVueVue
Setup difficultymoderatehardeasy
Complexity3/54/52/5
Audiencedeveloperdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Local mode needs your own LLM for some extraction features, cloud mode needs a ComPDF account and API key.

You can use and modify this freely, but if you distribute a modified version of the library itself, you must release those changes under the same license.

So what is it?

DocSlight is an open source tool that reads documents such as PDFs, scanned pages, images, and Office files like Word, PowerPoint, and Excel, and turns them into structured data you can use elsewhere, such as Markdown or JSON. It is part of a larger product family from the company behind ComPDF. Rather than just extracting raw text the way older OCR tools do, it combines AI-based parsing, text recognition for more than 80 languages with automatic language detection, and structured field extraction that can point back to exactly where on the page a value came from. The README lists its intended uses as building blocks for retrieval-augmented generation pipelines and knowledge bases, processing invoices, analyzing contracts and pulling out specific clauses, giving AI agents a tool to read documents, and broader enterprise document automation. There are two ways to run it. Local mode runs entirely on your own machine for free and needs no account, though some extraction features require you to supply your own local AI model. Cloud mode sends documents to the company's hosted service using an API key you generate after signing up, and offers higher accuracy plus a free usage quota. Either way you install it with pip and run a single parse command from the command line, or you can clone the repository and start a browser-based interface with Docker Compose that lets you drag and drop files. Beyond the open source Lite edition, the company also sells an Enterprise edition, available as a hosted service or a self-hosted deployment, that adds things like uploading from cloud storage or a scanner and batch processing. The project is released under the GNU LGPL v3 license. The full README is longer than what was shown.

Copy-paste prompts

Prompt 1
Install docslight with pip and parse this invoice.pdf into Markdown using local mode.
Prompt 2
Explain the difference between DocSlight's local mode and cloud mode based on its README.
Prompt 3
Set up DocSlight's Web UI with Docker Compose so I can drag and drop files to parse.
Prompt 4
Compare DocSlight's Lite and Enterprise editions for a document automation project.

Frequently asked questions

What is docslight?

An open source document parser that turns PDFs, scans, images, and Office files into structured Markdown or JSON using AI parsing and multilingual OCR, runnable locally or via a cloud API.

What language is docslight written in?

Mainly Vue. The stack also includes Python, Vue, Docker.

What license does docslight use?

You can use and modify this freely, but if you distribute a modified version of the library itself, you must release those changes under the same license.

How hard is docslight to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is docslight for?

Mainly developer.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.