Automation and scraping · Independent · 24 Oct 2026

Zipstack/unstract

LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows

← Automation and scraping

In simple English

This tool reads messy PDF files, scanned papers and photos of documents. You tell it what information you want in plain words. It uses artificial intelligence to find that information and turn it into neat digital tables. You can connect it to your database or run it as a web service. It runs on your own server and needs Docker (a way to run a program inside its own sealed box, so it cannot break anything else on your server).

What it is

Unstract is an open-source platform for extracting structured data from unstructured documents using language models and visual prompts. It combines a visual prompt studio, an API server, and automated data pipelines into one self-hosted system.

What it is good for

Solo builders who need to pull order details, invoices or medical forms into a database without writing custom regex patterns for every new document layout.

Stars7.3k
LicenceAGPL-3.0
Latestv0.196.0 (2026-10-23)checked 24 Oct 2026
Last code change2026-10-23
Written inPython
Setup effort
4 / 5
Payoff
5 / 5
Good
  • Build extraction rules using plain language prompts instead of fragile code patterns.
  • Pull files directly from cloud storage buckets and push clean results into SQL databases.
  • Connects with local models or major cloud providers for text processing.
Watch for
  • Needs Docker and at least 8 GB of memory to run the entire backend stack smoothly.
  • Complex multi-container setup requires patience for a non-programmer.
  • Licensed under AGPL-3.0, which has strict rules if you offer it as a hosted service to others.
Replaces

Writing custom regex parsers or paying expensive data extraction services

Works well with

n8n

Similar repositories
  • unclecode/crawl4ai86k stars

    Scrapes web pages for AI consumption, while Unstract handles static files and documents.

  • docling-project/docling69k stars

    Parses complex documents into structured formats locally, while Unstract provides an entire workflow platform with UI and pipelines.

Also reviewed on Repo Radar
  • apache/airflow47k stars

    Runs chains of scheduled jobs written in Python, and keeps a record of every run.

  • cybertec-postgresql/pg_timetable1.4k stars

    A job scheduler for PostgreSQL whose jobs and schedules live in the database itself.

  • scrapy/scrapy65k stars

    A Python framework for fetching and parsing large numbers of web pages.

Install
git clone https://github.com/Zipstack/unstract.git
cd unstract
./run-platform.sh

Independent: reviewed for Repo Radar outside the daily editions. Every independent repository

On Stack Radar: Commerce & Finance › Accounting & Expense Management › Financial Operations.

Get the next edition in your inbox

A dozen repositories, opened and checked. The licence read, the last release dated, and the ones that did not make it named with the reason. It is the half most lists leave out.

No tracking pixels. One click to leave. The archive stays free either way.

We use your address to send the edition and nothing else. Confirm by email, leave in one click. How we handle it.